341 24 214MB
English Pages [502] Year 2021
DISEASES OF THE NERVOUS SYSTEM
ELSEVIER science & technology books
Companion Website:
https://www.elsevier.com/books-and-journals/book-companion/9780128212288
Diseases of the Nervous System, Second Edition Harald Sontheimer Resources for Readers: Companion website provides the following resources: • Table of contents • About the author • Introduction • Figures from the volume in PowerPoint presentation
DISEASES OF THE NERVOUS SYSTEM SECOND EDITION Harald Sontheimer Harrison Distinguished Professor and Chair, Department of Neuroscience, University of Virginia, School of Medicine, Charlottesville, VA, United States
Academic Press is an imprint of Elsevier 125 London Wall, London EC2Y 5AS, United Kingdom 525 B Street, Suite 1650, San Diego, CA 92101, United States 50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom Copyright © 2021 Elsevier Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library ISBN 978-0-12-821228-8 For information on all Academic Press publications visit our website at https://www.elsevier.com/books-and-journals
Publisher: Nikki Levy Acquisitions Editor: Melanie Tucker Editorial Project Manager: Kristi Anderson Production Project Manager: Punithavathy Govindaradjane Cover Designer: Matthew Limbert Typeset by SPi Global, India
Dedication To my daughters Melanie and Sylvie, and to all the students who I taught over the past decades. You have served as a constant inspiration and motivated me to make Neuroscience more accessible to a broad readership.
This page intentionally left blank
Contents
About the Author Acknowledgments Introduction
II
xi xiii xv
4. Aging, Dementia, and Alzheimer Disease
I
HARALD SONTHEIMER
Static Nervous System Diseases 1. Cerebrovascular Infarct: Stroke HARALD SONTHEIMER
1 Case Story 3 2 History 4 3 Clinical Presentation/Diagnosis/Epidemiology 5 4 Disease Mechanism/Cause/Basic Science 8 5 Treatment/Standard of Care/Clinical Management 17 6 Experimental Approaches/Clinical Trials 20 7 Challenges and Opportunities 22 Acknowledgment 23 References 23
2. Central Nervous System Trauma HARALD SONTHEIMER
1 Case Story 2 History 3 Clinical Presentation/Diagnosis/Epidemiology 4 Disease Mechanism/Cause/Basic Science 5 Treatment/Standard of Care/Clinical Management 6 Experimental Approaches/Clinical Trials 7 Challenges and Opportunities Acknowledgments References
25 26 28 34 44 45 47 48 48
3. Seizure Disorders and Epilepsy
1 Case Story 81 2 History 82 3 Clinical Presentation/Diagnosis/Epidemiology 83 4 Disease Mechanism/Cause/Basic Science 87 5 Treatment/Standard of Care/Clinical Management 97 6 Experimental Approaches/Clinical Trials 102 7 Challenges and Opportunities 104 Acknowledgments 105 References 106
5. Parkinson Disease HARALD SONTHEIMER
1 Case Story 109 2 History 110 3 Clinical Presentation, Diagnosis, and Epidemiology 111 4 Disease Mechanism/Cause/Basic Science 115 5 Treatment/Standard of Care/Clinical Management 126 6 Experimental Approaches/Clinical Trials 129 7 Challenges and Opportunities 131 Acknowledgment 131 References 131
6. Diseases of Motor Neurons and Neuromuscular Junctions HARALD SONTHEIMER
HARALD SONTHEIMER
1 Case Story 2 History 3 Clinical Presentation/Diagnosis/Epidemiology 4 Disease Mechanism/Cause/Basic Science 5 Treatment/Standard of Care/Clinical Management 6 Experimental Approaches/Clinical Trials 7 Challenges and Opportunities Acknowledgments References
Progressive Neurodegenerative Diseases
51 52 53 57 69 74 75 75 75
1 Case Story 135 2 History 136 3 Clinical Presentation/Diagnosis/Epidemiology 137 4 Disease Mechanism/Cause/Basic Science 143 5 Treatment/Standard of Care/Clinical Management 154 6 Experimental Approaches/Clinical Trials 155 7 Challenges and Opportunities 158 Acknowledgments 158 References 158
vii
viii Contents
7. Huntington Disease
6 Challenges and Opportunities 254 Acknowledgments 254 References 254
HARALD SONTHEIMER
1 Case Story 161 2 History 162 3 Clinical Presentation/Diagnosis/Epidemiology 163 4 Disease Mechanism/Cause/Basic Science 166 5 Treatment/Standard of Care/Clinical Management 175 6 Experimental Approaches/Clinical Trials 176 7 Challenges and Opportunities 178 Acknowledgments 178 References 179
III Secondary Progressive Neurodegenerative Diseases 8. Multiple Sclerosis HARALD SONTHEIMER
1 Case Story 183 2 History 184 3 Clinical Presentation/Diagnosis/Epidemiology 186 4 Disease Mechanism/Cause/Basic Science 189 5 Treatment/Standard of Care/Clinical Management 198 6 Experimental Approaches/Clinical Trials 201 7 Challenges and Opportunities 204 Acknowledgments 204 References 204
9. Brain Tumors HARALD SONTHEIMER
1 Case Story 207 2 History 208 3 Clinical Presentation/Diagnosis/Epidemiology 210 4 Disease Mechanism/Cause/Basic Science 213 5 Treatment/Standard of Care/Clinical Management 225 6 Experimental Approaches/Clinical Trials 227 7 Challenges and Opportunities 231 Acknowledgments 232 References 232
10. Infectious Diseases of the Nervous System HARALD SONTHEIMER
1 Case Story 235 2 History 236 3 Clinical Presentation/Diagnosis/Epidemiology/Disease Mechanism 237 4 Beyond the Infection: Bona Fide Brain Disorders Involving Pathogens 251 5 Experimental Approaches/Clinical Trials 252
IV Developmental Neurological Conditions 11. Neurodevelopmental Disorders HARALD SONTHEIMER
1 Case Study 259 2 History 260 3 Development of Synapses in the Human Cortex and Diseases Thereof 261 4 Down Syndrome 263 5 Fragile X Syndrome 267 6 Rett Syndrome 270 7 Autism Spectrum Disorder (ASD) 272 8 Common Disease Mechanism 277 9 Challenges and Opportunities 278 Acknowledgment 278 References 278
V Neuropsychiatric Illnesses 12. Mood Disorders and Depression HARALD SONTHEIMER
1 Case Story 283 2 History 284 3 Clinical Presentation/Diagnosis/Epidemiology 285 4 Disease Mechanism/Cause/Basic Science 287 5 Treatment/Standard of Care/Clinical Management 295 6 Experimental Approaches/Clinical Trials 298 7 Challenges and Opportunities 300 Acknowledgment 300 References 300
13. Schizophrenia HARALD SONTHEIMER
1 Case Story 303 2 History 304 3 Clinical Presentation/Diagnosis/Epidemiology 305 4 Disease Mechanism/Cause/Basic Science 306 5 Treatment/Standard of Care/Clinical Management 318 6 Experimental Approaches/Clinical Trials 319 7 Challenges and Opportunities 323
Contents ix
Acknowledgments 323 References 323
14. Pain HARALD SONTHEIMER
1 Case Study 326 2 History 326 3 Clinical Presentation/Diagnosis/Epidemiology 327 4 Disease Mechanism/Cause/Basic Science 329 5 Common Forms of Pain and Currently Approved Treatments 339 6 Experimental Approaches/Clinical Trials 351 7 Challenges and Opportunities 354 Acknowledgments 354 References 354
3 Glutamate Toxicity 386 4 Protein Aggregates and Prion-Like Spread of Disease 391 5 Mitochondrial Dysfunction 393 6 Heritability of Disease With Elusive Genetic Causes 395 7 Epigenetics 398 8 Noncell Autonomous Mechanisms 401 9 Inflammation 402 10 Vascular Abnormalities 407 11 Brain-Derived Neurotrophic Factor 408 12 Challenges and Opportunities 412 References 412
VII Bench-To-Bedside Translation
15. Drug Addiction
17. Drug Discovery and Personalized Medicine
HARALD SONTHEIMER
HARALD SONTHEIMER
1 2 3 4
1 Introduction 417 2 How Did We Get to This Point? A Brief History 417 3 Drug Discovery: How Are Candidate Drugs Identified? 418 4 What Are Clinical Trials and Why Do Them? 420 5 The Placebo Effect 424 6 Why Do Clinical Trials Fail? 425 7 Personalized Medicine 427 8 Challenges and Opportunities 427 References 428
Case Story 357 History of Drug Use and Addiction 358 Biology of Substance Use and Addiction 361 Common Substance Use Disorders, Underlying Biology, Epidemiology, and Treatment 368 5 Challenges and Opportunities 379 Acknowledgments 379 References 379
VI
VIII
Common Concepts in Neurological and Neuropsychiatric Illnesses
Neuroscience Jargon
16. Shared Mechanisms of Disease
18. “Neuro”-Dictionary
HARALD SONTHEIMER
HARALD SONTHEIMER
1 Introduction 2 Neuronal Death
386 386
Index 469
This page intentionally left blank
About the Author Dr. Sontheimer is a researcher and educator with a lifelong interest in neuroscience. A native of Germany, he obtained a Master’s degree in evolutionary comparative neuroscience from the University of Ulm in which he worked on the development of occulomotor reflexes. In 1989, he obtained a doctorate in biophysics and cellular & molecular neuroscience from the University of Heidelberg, studying biophysical changes that accompany the development of oligodendrocytes, the principal myelinating cells of the nervous system. He moved to the United States, where he later became a citizen, for postdoctoral studies at Yale University. His independent research career began at Yale in 1991 and continued at the University of Alabama Birmingham during 1994–2015, and, more recently, at Virginia Tech and the University of Virginia. His research focuses on the role of glial support cells in health and disease. His laboratory has made major discoveries that led to two clinical trials using novel compounds to treat malignant gliomas. His research led to over 190 peer-reviewed publications. For the clinical development of his discoveries, Dr. Sontheimer started a biotechnologies company, Transmolecular Inc., which conducted both phase I and phase II clinical trials with the anticancer agent, chlorotoxin. Morphotec Pharmaceuticals, which will be conducting the phase III clinical trials, recently acquired this technology. As educator, Dr. Sontheimer has been active in teaching medical neuroscience, graduate cellular and molecular neuroscience, and, for the past 10 years, he has offered both graduate and undergraduate courses on diseases of the nervous system. In 2005, Dr. Sontheimer became director of the Civitan International Research Center, a philanthropically
supported center in Birmingham AL devoted to the study and treatment of children with developmental disabilities, ranging from Down’s syndrome to autism. In this capacity, Dr. Sontheimer was frequently tasked with explaining complex scientific processes to a lay audience. Recognizing the need to further educate the public about neurological disorders using language that is accessible to an educated public motivated Dr. Sontheimer to write a textbook on diseases of the nervous system. To ensure that the material is comprehensive yet readily understandable, he wrote large parts of this text while on sabbatical leave at Rhodes College in Memphis, where he taught undergraduates while testing his book on this group of talented third- and fourth-year neuroscience students. In 2015, Dr. Sontheimer was tapped to found a school of neuroscience at Virginia Tech with the goal to offer a unique neuroscience education to an increasing number of undergraduates. As the first of its kind, this enterprise devoted an entire school to a variety of neuroscience experiences that include majors in clinical, experimental, cognitive, systems, computational, and social neuroscience. In 2020, Dr. Sontheimer was recruited to the University of Virginia School of Medicine as the Chair of Neuroscience with the mission to build this department into a leading research enterprise devoted to discovery and translation science in neuroimmunology and neurodegenerative diseases. Dr. Sontheimer continues to manage a very active research laboratory where he involves a spectrum of trainees ranging from undergraduates to postdoctoral scientists. Dr. Sontheimer has trained over 50 PhD and MD/PhD students and postdoctoral fellows, many of whom have independent faculty positions today.
xi
This page intentionally left blank
Acknowledgments English is a second language for me. To make up for my shortcomings, I am indebted to several colleagues who have meticulously reviewed every word I wrote. Foremost, my long-term Assistant, Anne Wailes, who tirelessly edited and polished every sentence in the first edition of this book. She also tracked down the copyrights for hundreds of figures that were reproduced in this book. Anne did all this while attending to the many daily tasks of administrating a large research center and looking after my trainees in my absence. This was a monumental undertaking and words cannot describe how fortunate I feel to have had her support throughout this journey. In addition, each chapter went through two stages of scientific review. The first stage of review was conducted by two colleagues to whom I am tremendously indebted. The first edition was reviewed for scientific content and accuracy by a tremendously talented postdoc, Dr. Alisha Epps, who, for an entire year, spent almost every weekend reading and correcting book chapters as I completed them. Alisha had a talent to simplify and clarify many difficult concepts, and, if needed, she found suitable figures or even drew them from scratch. The second edition was reviewed by my colleague and friend Dr. Kristin Phillips. As collegiate Professor in Neuroscience, she is an equally enthusiastic reader of Neuroscience literature and had developed a study abroad course that examines cultural and societal difference in the application of Neuroscience to Medicine. Co-teaching this course entitled “Global Perspectives in Neuroscience” I realized that my chapters must take a more global look at disease epidemiology and consider discrepancies in disease presentations and outcomes. Her contributions to this book were tremendous and I am indebted to her generous support. The second stage of review involved experts in the respective disease. I am privileged to have a number of friends who are clinicians or clinician–scientists and who were willing to selflessly spend countless hours correcting the mistakes I had made. While I am acknowledging each person with the very chapter they reviewed, I like to acknowledge all of them in this introduction by name. Alan Percy, MD, PhD, University of Alabama Birmingham Amie Brown McLain, MD, University of Alabama Birmingham Anthony Nicholas, MD, PhD, University of Alabama Birmingham
Christopher B. Ransom, MD, PhD, University of Washington Erik Roberson, MD, PhD, University of Alabama Birmingham James H. Meador-Woodruff, MD, University of Alabama Birmingham Jeffrey Rothstein, MD, PhD, Johns Hopkins University Leon Dure, MD, University of Alabama Birmingham Louis Burton Nabors, MD, University of Alabama Birmingham Richard Sheldon, MD, University of Alabama Birmingham Stephen Waxman, MD, PhD, Yale University Steven Finkbeiner, MD, PhD, The Gladstone Institute for Neurological Disease Thomas Novack, PhD, University of Alabama Birmingham William Britt, MD, University of Alabama Birmingham Warren Bickel, PhD, Virginia Tech To be able to spend a year and a half writing a book is a luxury and privilege that, even in academia, only a few people enjoy. The first edition was developed while I was still at the University of Alabama at Birmingham. I like to thank the Dean, President, and my Chairman for enthusiastically supporting this endeavor. During the spring semester of 2014, I became a visiting Professor, embedded among the wonderful faculty of Rhodes College in Memphis TN, a picturesque small liberal arts college. I am thankful for the hospitality and support of all the Rhodes administrators and faculty, many of whom I engaged in inspirational discussion during lunch or coffee breaks. I am particularly grateful to their Neuroscience program for letting me participate in their curriculum and take residence in Clough Hall. The writing of the second edition accompanied my building the School of Neuroscience at Virginia Tech, which, over the course of 5 years grew to be one of the largest undergraduate programs in the country. Here I took on several undergraduate courses ranging from Neuroscience of the Mind, Brain and Pain, to my flagship course for which this book was written: Disease of the Nervous System. Throughout, many students provided invaluable feedback on this book, some formal, using a prescribed feedback form, other informal during office hours. I am thankful to the many students who attended
xiii
xiv Acknowledgments my classes at Rhodes, UAB, and Virginia Tech and who took a particular interest and regularly provided recommendation for improvements. I trust that many of them are either in Graduate or in Medical school by now, and I wish them well.
My final acknowledgment goes to my publisher, Elsevier Academic Press, for their tremendous work editing, publishing, and marketing this book. Particularly to the editorial project manager Kristi Anderson, the senior acquisitions editor Melanie Tucker, and their production team.
Introduction The study of nervous tissue and its role in learning and behavior, which we often call neuroscience, is a very young discipline. Johannes Purkinje first described nerve cells in the early 1800s, and by 1900, the pathologist Ramón y Cajal generated beautifully detailed histological drawings illustrating all major cell types in the brain and spinal cord and their interactions. Cajal also described many neuron-specific structures including synaptic contacts between nerve cells; yet how these structures informed the brain to function like a biological computer remained obscure until recently. Although Luigi Galvani’s pioneering experiments in the late-1700s had already introduced the world to biological electricity, ion channels and synaptic neurotransmitter receptors were only recognized as “molecular batteries” in the late-1970s and early 1980s. The first structural image of an ion channel was generated even more recently in 1998, and for many ion channels and transmitter receptors, such information still eludes us. Surprisingly, however, long before neuroscience became a freestanding life science discipline, doctors and scientists had been fascinated with diseases of the nervous system. Absent any understanding of cellular mechanisms of signaling, many neurological disorders were quite accurately described and diagnosed in the early to mid-1800s, including epilepsy, Parkinson Disease, schizophrenia, multiple sclerosis, and Duchenne muscular dystrophy. During this period and still today, the discovery process has been largely driven by a curiosity about disease processes. What happens when things go wrong? Indeed, much of the early mapping of brain function was only possible because things went very wrong. Had it not been for brain tumors and intractable epilepsy, surgeons such as Harvey Cushing and Wilder Penfield would have had no justification to open the human skull of awake persons to establish functional maps of the cortex. Absent unexpected consequences of surgery, such as the bilateral removal of the hippocampi in HM that left him unable to form new memories, or unfortunate accidents exemplified by the railroad worker, Phineas Gage, who destroyed his frontal lobe in a blast accident, we would not have had the opportunity to learn about the role of these brain structures in forming new memories or executive function, respectively. Such fascination with nervous system disease and injury continues to date, and it is probably fair to say that neuroscience is as much a study of health as that of disease.
For the past 20 years, I have been teaching a graduate course entitled “Diseases of the Nervous System” and more recently, I added an undergraduate course on the same topic as well. Every year, almost without fail, students would ask me whether I could recommend a book that they could use to accompany the course. I would usually point them to my bookshelf, filled with countless neuroscience and neurology textbooks ranging from Principles in Neuroscience to Merritt’s Neurology. When I started this book project, there was indeed no such book, yet I hoped that sooner or later some brave neuroscientist would venture to write a book about neurological illnesses. Surprisingly, this did not happen, so in 2014, I decided to fill this void. My initial inclination was to produce a multiauthor edited book. By calling on many friends and colleagues to each write a chapter on their favorite disease, this should be a quick affair. However, from own experience, I knew that book chapters are always the lowest priority on my “to do” list, and I really was eager to pester my colleagues monthly to deliver their goods. Ultimately, they would surely ask a senior postdoc to take the lead and in the end, the chapters would be heterogeneous and not necessarily at a level appropriate for a college audience. For my target audience, this book needed to be a monograph. While I did not know at the time what I was getting into, I spent the majority of 2014 and 2015 reading over 2500 scientific papers and review articles while also writing for about 7–10 h daily. I felt exhausted yet also became quite a bit more educated in the process. Given the rapid progress in research and discovery, 5 years later, in 2019–20, I repeated this exercise and wrote this second edition, which includes major updates and new additional chapters on Pain and Addiction. The target audience for this book is any student interested in neurological and neuropsychiatric illnesses. This includes undergraduates, early graduate students, and medical students taking a medical neuroscience course. I also expect the material to be of benefit to many health professionals who are not experts in the field. The book may even appeal to science writers or simply a science-minded layperson, possibly including persons affected by one of the illnesses. Purposefully, the book lacks a basic introduction to neuroscience as I would expect the reader to have a basic understanding of neurobiology. Many excellent textbooks have been written, each of which would prepare one well to comprehend
xv
xvi Introduction this text. I feel that I could not have done justice to this rapidly expanding field had I attempted to write a short introduction. However, to at least partially make up for this, I include an extensive final chapter that is called “Neuroscience Jargon.” I consider this more than just a dictionary. It has a succinct summary of approximately 500 of the most important terms and is written as nontechnically as possible. I hope that this will assist the reader to get his/her bearings as needed. The book makes every effort to cover all the major neurological illnesses that affect the central nervous system though it is far from complete. My intention was to go fairly deep into disease mechanisms and this precluded a broader coverage of small and less well-known conditions. I found it useful to group the diseases into five broad categories that provided some logical flow and progression. Specifically, I begin with static illnesses, where an acute onset causes immediate disability that typically does not worsen over time. This group is best exemplified by stroke and CNS trauma but also includes genetic or acquired epilepsy (Chapters 1–3). I next covered the classical primary progressive neurodegenerative diseases including Alzheimer, Parkinson, Huntington, and ALS (Chapters 4–7). For each of these chapters, I added some important related disorders. For example, the chapter on Alzheimer includes frontal temporal dementia; for Parkinson, I included essential tremors and dystonia, and for Huntington, I touch on related “repeat disorders” such as spinocerebellar ataxia. The chapter that covers ALS includes a variety of disease along the motor pathway essentially moving from diseases affecting the motor neurons themselves (ALS), their axons (Guillain-Barre syndrome), to the presynaptic (Lambert Eaton myotonia), and postsynaptic (myasthenia gravis) neuromuscular junction. Next, I progressed to neurodegenerative diseases that are secondary to an insult yet still cause progressive neuronal death. I call these secondary progressive neurodegenerative diseases and the examples I am covering include multiple sclerosis, brain tumors, and infections (Chapters 8–10). It may be unconventional to call these secondary neurodegenerative diseases yet in multiple sclerosis, the loss of myelin causes progressive axonal degeneration, brain tumors cause neurological symptoms by gradually killing neurons, and infection causes progressive illnesses again by progressively killing neurons. Nervous system infection could have quickly become an unmanageable topic since far too many pathogens exist that could affect the nervous system. I therefore elected to discuss important examples for each class of pathogen (prion proteins, bacteria, fungi, viruses, single- and multicellular parasites). While none of these pathogens are brain-specific, I chose examples in which the nervous system is primarily affected including meningitis, botulism, tetanus, poliomyelitis, neurosyphilis, brain-eating
amoeba, neurosistercosis, neuroaids, and prion diseases. I also used this chapter as an opportunity to highlight the tropism displayed by some viruses for the nervous system and how this can be harnessed to deliver genes to the nervous system for therapeutic purposes. For the section on neurodevelopmental disorders, I similarly chose four important examples including Down syndrome, Fragile X, autism, and Rett syndrome. These disorders have so many commonalities that it made sense to cover them in a single chapter (Chapter 11). No contemporary book of nervous system disease would be complete without coverage of neuropsychiatric illnesses and I elected to devote one chapter each to depression (Chapter 12) and schizophrenia (Chapter 13). Finally, for the second edition, I also included pain (Chapter 14) and addiction (Chapter 15), two topics that intersect on the pervasive issue of addiction to opioid pain killers. Taken together, I believe the material covers the “big” brain disorders that any neuroscientist or medical student should know. However, anyone looking for more detailed information on rare disorders or disorders primarily affecting the peripheral nervous system or sensory organs is referred to some of the excellent neurology textbooks that I cite as my major sources throughout the book. To ensure that the material is presented in an accessible, yet comprehensive format, the book was developed in a uniquely student-centered way, using my target audience as a focus group. To do so, I wrote the book as accompanying text to an undergraduate course, writing each chapter as I was teaching to neuroscience majors. The first edition was written while on sabbatical leave at Rhodes College in Memphis TN, a small and highly selective Liberal Arts college. The second edition I wrote at Virginia Tech, where I moved in 2015 to build the School of Neuroscience, an entire School devoted to Neuroscience. Each week, I handed out a new disease chapter, and after giving a 75-min lecture, small groups of students had to prepare independent lectures that they delivered to the class based on recent influential clinical and basic science papers that I assigned (and list in this book with each chapter). Each week, using a questionnaire, the students provided detailed feedback on how accessible, interesting, and complete my chapters were, and how well the book prepared them for the assigned papers that they had to present in class. I took their comments very seriously, frequently spending days incorporating their suggestions. I am thankful to all of them, as it made the book a better read. As I began my research, a challenge that became immediately evident was the sheer magnitude of the available literature. Moreover, writing about a disease that is outside ones’ personal research specialty leaves one without a compass to decide which facts are important
Introduction xvii
and which are not. Narrowing literature searches to just “diseases” and “review articles” did not help much and only marginally reduced the number of hits from the tens of thousands into the thousands. While it was gratifying to see the enormous amount of information that has been published, it was daunting to filter and condense this material into a manageable number of sources. In the end, I developed a strategy to first identify the “opinion leaders” in each field, and then, using their high-impact reviews, widen my search to include reviews that appeared to cover the most salient points on which the entire field appears to largely agree upon, while staying largely out of more tentative emerging and controversial topics. This was important since the objective of this textbook was to introduce current accepted concepts rather than speculations. Another challenge I faced was to keep the material interesting. As teacher of medical neuroscience, I have long recognized the value of clinical cases. I decided to start each disease chapter with case story, which is either an actual case or one close to cases that I have actually witnessed in some form or other. The students liked this format, particularly since many of the cases I describe involve young people. To offer perspective on each disease, I also elected to provide a brief historic review for each disease. How long has society been dealing with stroke, epilepsy, Huntington, or autism? What were early interpretations on the disease cause, how was disease treated, and what were the most informative milestones? This was possibly the hardest section for me to write, since good sources were difficult to find. Yet it was also the most fascinating. The students initially had little appreciation for these sections and really did not see much value in them. However, this changed after we discussed the value of what I call “science forensics” and the historic insight that could be gleaned. We discussed how the history of disease, when viewed in the context of the history of mankind, allows us to dismiss or consider human endeavors and exposure to man-made chemicals as disease causes. After we discussed how Mexican vases made over 600 years ago already depicted children with Down syndrome, or how Polio crippled children were portrayed on Egyptian stilts that were over 2000 years old, it became clear that neither of these conditions was modern at all. Historic accounts similarly suggest that environmental exposures are unlikely contributors to stroke or epilepsy. Yet, by contrast, the earliest accounts of Parkinson Disease align perfectly with the early industrial revolution of the mid-1800, making industrial pollutants potential disease contributors. Even more extreme, no historic account for autism exists prior to the 1930s. Clearly, for some of those diseases, human influences must be considered as contributory factors. The historic adventures also allowed me to examine diseases in the context of society at a given time in h istory,
clearly important lessons when teaching neuroscience at a Liberal Arts college. Our classes included how patients with epilepsy were labeled witches and burned in medieval Europe; how the heritability of diseases such as Huntington corrupted even doctors to subscribe to the reprehensible teachings of the eugenics movement; or how the infamous Tuskegee syphilis studies served as the foundation for the protection of human subjects participating in human clinical trials, measures that we take for granted today. Another lesson learned from the ancient accounts of Down syndrome is that childbirth late in a mother’s life occurred throughout history, but more importantly that those children were cared for in many societies with the same love and compassion we have for them today. The majority of pages in this book are devoted to the biology of each disease. It is remarkable how much we know and how far we have come in just the past few decades, from the historic disease pathology-focused approach to contemporary considerations of genes and environmental interactions causing disease in susceptible individuals. It is fascinating to note how cumbersome the initial positional cloning efforts were that identified the first candidate genes for disease compared to today’s large genome-wide association studies that identify large networks of gene and their interactions. Clearly, we experience a transformational opportunity to study and understand disease through the study of rare genetic forms of familial diseases that can inform us about general disease mechanisms and allow us to reproduce disease in genetic animal models. At the same time, it is sobering to see how often findings in the laboratory fail to subsequently translate into better clinical practice. I devote a considerable amount of discussion to such challenges and end each chapter with a personal assessment of challenges and opportunities. After completing the disease chapters, it was clear that there were many cross-cutting shared mechanisms and features of neurological disease that I elected to devote an entire chapter solely to shared mechanisms of neurological illnesses (Chapter 16). Not surprisingly, almost all the class discussions sooner or later gravitated toward ways to translate research findings from the bench to the bedside. Yet few of the students had any idea what this really entails or the challenges that clinical trials face. Having been fortunate enough to develop an experimental treatment for brain tumors in my laboratory that I was able to advance from the bench into the clinic through a venture capital-supported biotech startup, I felt well equipped to discuss many of the challenges in proper perspective. So I devoted an entire chapter (Chapter 17) to this important, albeit not neuroscience-specific, topic. The class included important discussions on the placebo effect and frank conversations as to why many scientific findings cannot be reproduced, and why most clinical trials ultimately fail.
xviii Introduction I also added several provocative topics to class discussions such as the questionable uses of neuroscience in marketing and advertising and the controversial use of neuroscience in the courtroom. Since neither relates to specific neurological diseases, I elected to leave this out of the book but encourage neuroscience teachers to bring such topics into the classroom as well. One thing that troubled me throughout my writing was the way in which sources are credited in textbooks. As a scientist, I reflexively place a source citation behind every statement I make. In the context of this book, however, I could only cite a few articles restricting myself to ones that I felt were particularly pertinent to a given statement. A list of general sources that most informed me in my reading is included at the end of each chapter. I am concerned, however, that I may have gotten a few facts wrong, and that some of my colleagues will contact me, offended that I ignored one of their findings that they consider ground breaking; or if I mentioned them, that I failed to explicitly credit them for their contribution. It was a danger that I had to accept, albeit with trepidation and I hope that any such scientists will accept my preemptive apologies. To mitigate against factual errors, I reached out to many colleagues around the country, clinical scientists whom I consider experts in the respected disease, and asked them to review each chapter. I am indebted to these colleagues, whom I credit with each chapter, who selflessly devoted many hours to make this a better book. Their effort has put me at greater ease and hopefully will assure the reader that this book represents the current state of knowledge.
Given that the book was developed as an accompaniment to a college course, I expect that it may encourage colleagues to offer a similar course at their institution. I certainly hope that this is the case. To facilitate this, I am happy to share PowerPoint slides of any drawings or figures contained in this book, as well as any of the 1000 + slides that I made to accompany this course. I can be contacted by email at [email protected]. Also, for each chapter I am listing a selection of influential clinical and basic science articles that I used in class. These are just my personal recommendations and not endorsements of particular themes or topics. These papers have generated valuable discussion and augmented the learning provided through the book. Finally, as I finish editing the second edition of this book, in which I incorporated many scientific advances and clinical trials occurred since the first publication in 2015, I still keep finding more and more articles reporting exciting new scientific discoveries that I would have liked to include. However, if I did, this book would never reach the press. It is refreshing to see that neuroscience has become one of the hottest subjects in colleges and graduate schools and even the popular press. Neuroscience research is moving at a lightning pace. It is therefore unavoidable that the covered material will only be current for a brief moment in time, and, as you read this book, that time will have already passed. Harald Sontheimer
S E C T I O N
I
STATIC NERVOUS SYSTEM DISEASES
This page intentionally left blank
C H A P T E R
1 Cerebrovascular Infarct: Stroke Harald Sontheimer O U T L I N E 1
Case Story
3
2 History
4
3
Clinical Presentation/Diagnosis/Epidemiology
5
4
Disease Mechanism/Cause/Basic Science 4.1 Causes of Vessel Occlusions: The Thrombolytic Cascade 4.2 The Ischemic Cascade 4.3 The Ischemic Penumbra 4.4 The NMDA Receptor and Glutamate Excitotoxicity 4.5 Role of Glutamate 4.6 NMDA Inhibitors to Treat Stroke 4.7 Effect of Temperature 4.8 Stroke Genetics
8
5
Treatment/Standard of Care/Clinical Management 5.1 Chemical Thrombolysis Using Intravenous tPA 5.2 Mechanical Thrombolysis by Endovascular Therapy for Ischemic Stroke
6
10 11 13 15 15 15 16 17
7
17 17 18
1 CASE STORY
19 19 19 19
Experimental Approaches/Clinical Trials 6.1 Neuroprotection 6.2 Hypothermia 6.3 Improved Clot Busters 6.4 Brain Rewiring After Stroke 6.5 Why Have So Many Promising Drugs Failed in Human Clinical Trials?
20 20 21 21 21
Challenges and Opportunities
22
22
Acknowledgment
23
References
23
General Readings Used as Source
24
Suggested Papers or Journal Club Assignments Clinical Papers Basic Papers
24 24 24
get up and take some ibuprofen. However, getting out of bed turned into a struggle. She did not sense her right hand and could not move her right leg. As she rolled toward the edge of the bed, her vision blurred. Her eyes felt like they were pushing out of her head. “Where is the phone?” She used her left hand to scan her night stand one handspan at a time. Once her hand made contact with her cell phone, she struggled to recognize the screen. She was panicking. Who to call? Amy, her best friend should be up by now. After 20 rings, Amy finally answered. “Why up so early? I am sleeping.” “Amy, I can’t move, I am trapped in my bed with a brutal headache and I can’t see well.” It didn’t take Amy long to realize that her friend was in serious trouble. Amy had
Natalie was excited to start her senior year at Virginia Tech. She saved some of the most interesting art history and creative writing classes for her last year. She was equally excited to participate one last season in Cheerleading for the Hokies football team. She still gets a rush by the pregame pageantry and a stadium trembling as the players enter the stadium to the roaring sounds of “Enter Sandman.” This Saturday morning, she woke unusually early and was not feeling well at all. Her head hurt and although she had been partying the night before, this did not feel like a hangover headache. After rolling in pain for a few minutes, she decided to
Diseases of the Nervous System. https://doi.org/10.1016/B978-0-12-821228-8.00001-9
5.3 Anticlotting Factors to Prevent Recurrence 5.4 Treatment of Hemorrhagic Stroke 5.5 Rehabilitation 5.6 Stroke Prevention
3
Copyright © 2021 Elsevier Inc. All rights reserved.
4
1. Cerebrovascular Infarct: Stroke
been volunteering for the VT Rescue Squad for the past 2 years and had ran several codes quite similar to this one. But never in a person Natalie’s age! She rushed to Natalie’s apartment while calling VT rescue on her way. They arrived just a minute apart and the team pried open the door with force only to find Natalie next to her bed crying unconsolably. At this point, she was barely responding to the rescue team. Her right face was drooping, and her arm was limb. Realizing that this may be a stroke, the team called ahead to Louis Gale Hospital to have the emergency room ready. Everything was a blur. Natalie was lifted on the gurney and quickly carried to the ambulance where Amy jumped in next to her, holding her friends hand all the way. The 6-min drive seemed like ages, and by the time they arrived, Natalie was no longer responding to Amy calling her name. Without delay, Natalie was moved to the imaging center for a CT scan. Low and behold, Natalie had near complete loss of blood flow in her left brain, particularly the central part. Ischemic stroke is most likely caused by an embolus or thrombus in the middle cerebral artery. This occludes blood flow to the most important parts of the brain controlling sensation and movement of the right body as well as speech and language. “Who has last seen her responsive?” asked the emergency room physician. Thankfully Amy was there to describe the events this morning. “How long ago did she call you?” “About 50 minutes ago,” Amy answered. As imaging has ruled out a hemorrhage, and Natalie’s blood pressure was 123/78, the emergency team decided to deliver a bolus injection of tPA, a clot-busting chemical, and hooked her up to a continuous infusion to deliver more of the drug. The next hour, however, did not yield any improvement and a subsequent CT scan showed little change. “Get her to Roanoke for surgical thrombectomy,” ordered the attending neurologist, “and use the helicopter to get there fast.” Less than 30 min later, Natalie arrived on the helipad of Roanoke Memorial Hospital and was greeted by a medical team that would take her to the angio suite where the radiologist threaded a catheter up her femoral vein toward the head and into the MCA. Once he navigated to the blocked artery, he actuated a small mesh that encapsulated the thrombus and slowly began to pull back. Within no time, blood flow returned, and Natalie started talking. Almost miraculously, she was able to move her right arm and her speech, while still slurred, was intelligible. Two hours later, Natalie was talking to Amy in the recovery room, and she was seriously considering going to the football game that evening. While that obviously did not happen, Natalie was back home 2 days later on a prescription of warfarin, a blood thinning medication. Had it not been for Amy quickly recognizing her friend’s condition and the proximity to a level one trauma center with a skilled interventional radiologist, Natalie may
have permanently lost movement of her right body or worse, may have died. She was among the 1 out of every 20 patients who were able to benefit from recent medical advances in stroke management using a combination of chemical and mechanical clot busters. During her medical follow-up, it was determined that she suffered from a congenital heart condition, a patent foramen ovale, in which the left and right atria of the heart are connected by a hole. This probably allowed a thrombus from her lower legs to find its way into the cerebral circulation rather than being filtered out by passing through the lungs. On the recommendation of her cardiologist, Natalie had heart surgery to close the foramen ovale 3 months after her stroke. This should lower her risk for stroke recurrence to that of the general population.
2 HISTORY Without recognizing its underlying cause, Hippocrates (460 BC), the “father of medicine,” provided the first clinical report of a person being struck by sudden paralysis, a condition he called apoplexy. This Greek word, meaning “striking away,” refers to a sudden loss of the ability to feel and move parts of the body and was widely adopted as a medical term until it was replaced by cerebrovascular disease at the beginning of the 20th century. Most patients and the general public prefer the term stroke, which first appeared in the English language in 1599. It conveys the sudden onset of a seemingly random event. Hippocrates explained apoplexy using his humoral theory, according to which the composition and workings of the body are based on four distinct bodily fluids (black bile, yellow bile, phlegm, and blood), which determine a person’s temperament and health. Diseases result from an imbalance in these four humors, with apoplexy specifically affecting the flow of humors to the brain. Humors were rebalanced through purging and bloodletting, which became the treatment of choice for stroke throughout the middle ages. The first scientific evidence that a disruption of blood flow to the brain causes stroke came through a series of autopsies conducted by Jakob Wepfer in the mid-1600s, yet the humoral theory of Hippocratic medicine ruled until the German physician Rudolf Virchow discredited it in his “Theories on Cellular Pathology,” published in 1858. Virchow, who made countless impactful contributions to medicine, was the first to explain that blood clots forming in the pulmonary artery can cause vascular thrombosis, and fragments arising from these thrombi can enter the circulation as emboli. These emboli are carried along with blood into remote blood vessels, where they can occlude blood flow or rupture vessels. His theory was initially based only on patient autopsies. However, together with
I. STATIC NERVOUS SYSTEM DISEASES
3 Clinical presentation/diagnosis/epidemiology
his student, Julius Cohnheim, Virchow went on to test this idea by injecting small wax particles into the arteries of a frog’s tongue to show that the wax acted as an embolus that shut off blood flow to the parts of the tongue supplied by this vessel. In subsequent studies, Cohnheim showed that an embolus can cause either blockade (ischemic stroke) or rupture (hemorrhagic stroke), contradicting competing views at the time that suggested that only blood vessel malformations or aneurisms could hemorrhage. It is worth noting that in the early 20th century, the recognition that emboli cause the selective abolition of blood flow in cortical blood vessels gave neurologists the first insight into functional neuroanatomy, showing selective and predictive deficits in sensory and motor function depending on where an embolus occluded a vessel. Throughout history and well into the 19th century, it was common to view stroke as a divine intervention, a summons to duty. Stroke was regarded as God’s punishment for unacceptable behavior. Shockingly, in spite of Virchow’s discoveries on thromboembolism, even major medical textbooks continued to blame the patient for the disease. For example, Osler’s medical textbook (1892) suggested that “the excited action of the heart in emotion may cause a rupture.” Others even suggested that a patient’s physical attributes, namely, a short, thick neck, and a large head, were predisposing factors.1 Interestingly, a diet “high in seasoned meat, poignant sauces and plenty of rich wine” was already accurately predicted by Robinson as a stroke risk factor in 1732.1, 2 We have obviously come a long way in the past 100 years. The routine medical use of X-rays, introduced in 1895, ultimately led to the development of the now widely available computed tomography (CT), with which it is possible to quickly and accurately localize blood clots or bleeds to guide further intervention. Surgical, mechanical, or chemical recanalization are now standard procedures, and various forms of image-guided stenting procedures, adopted from cardiac surgery, are now able to open cerebral vessels. The discovery of tissue plasminogen activator (tPA), approved as a chemical “clot buster” in 1996, was a major advance in the clinical management of acute stroke. Together with widely adopted rehabilitation, the outlook for many stroke patients has improved considerably.
3 CLINICAL PRESENTATION/ DIAGNOSIS/EPIDEMIOLOGY Cerebrovascular infarct is defined by the sudden onset of neurological symptoms as a result of inadequate blood flow. This is commonly called a stroke because the disease comes on as quickly as a “stroke of lightning” and without warning; we use the terms stroke and cerebrovascular infarct interchangeably throughout. We typically distinguish
5
three major stroke types that differ by their underlying cause and presentation. Focal ischemic strokes make up the vast majority of cases (∼ 80%) and result from vessel occlusion by atherosclerosis or blockage by an embolus or thrombus that causes a focal neurological deficit. Global ischemic strokes, often called hypoxic-ischemic injury, are more rare (10%) and result from a global reduction in blood flow, for example, through cardiac insufficiency. The neurological deficit affects the entire brain and is typically associated with a loss of consciousness. Finally, hemorrhagic strokes result from rupture of fragile blood vessels or aneurysm. These account for 10% of all strokes and can present with focal deficits if a small vessel is affected or global deficits if massive intracranial bleeding occurs. For the majority of patients who suffer an ischemic stroke, maximal disability occurs immediately after the blockage forms, without further worsening unless secondary intracranial bleeding occurs. Once the obstruction clears, the patient’s symptoms improve. However, a stroke patient has a greatly increased likelihood of recurrence: 20–30% of patients experience a second stroke within a year after the first insult. Hemorrhagic stroke is a severe medical emergency, with mortality approaching 40%. Symptoms are often progressive as bleeding continues, and a loss of consciousness is common. Neurological symptoms of stroke vary depending on the brain region affected; most strokes are focal and affect only one side of the body with muscle weakness and sensory loss. Telltale signs (Table 1) include a drooping face, change in vision, inability to speak, weakness and sensory loss (preferentially on one side of the body), and severe sudden-onset headaches. Many of these symptoms clear once blood flow to the affected brain region is restored. Therefore, rapid diagnosis and immediate medical intervention are of the essence, and anyone suspected of suffering a stroke should immediately call for emergency medical services. The popular catchphrase: “Time lost is TABLE 1 Common symptoms of a focal stroke. Alteration in consciousness; stupor or coma, confusion or agitation/ memory loss seizures, delirium Headache, intense or unusually severe often associated with decreased level of consciousness/neurological deficit, unusual/severe neck or facial pain Aphasia (incoherent speech or difficulty understanding speech) Facial weakness or asymmetry, paralysis of facial muscles (e.g., when patients speak or smile) Incoordination, weakness, paralysis, or sensory loss of one or more limbs (usually one half of the body and in particular the hand) Ataxia (poor balance, clumsiness, or difficulty walking) Visual loss, vertigo, double vision, unilateral hearing loss, nausea, vomiting, photophobia, or phonophobia
I. STATIC NERVOUS SYSTEM DISEASES
6
1. Cerebrovascular Infarct: Stroke
TABLE 2 Act FAST emergency response issued by the National Stroke Association. Use the FAST test to remember warning signs of stroke F = FACE
Ask the person to smile. Does one side of the face droop?
A = ARMS
Ask the person to raise both arms. Does one arm drift downward?
S = SPEECH
Ask the person to repeat a simple sentence. Does the speech sound slurred or strange?
T = TIME
If you observe any of these signs (independently or together), call 9-1-1 immediately.
FIGURE 2 Representative examples of computed tomography
brain lost,” highlights the sense of urgency. In the case of hemorrhaging stroke, symptoms may worsen rapidly because intracranial bleeding affects vital brain functions, and patients may lose consciousness. To encourage rapid admission of potential stroke victims to a hospital, the National Stroke Association devised the Face Arm and Speech Test (FAST), which aids the public in quickly identifying the major warning signs of the disease (Table 2). Once the patient is receiving medical care, a diagnostic decision tree is typically followed to guide treatment, as illustrated in Figure 1. Immediately upon admission to a hospital, the distinction between ischemic (occluding) stroke and hemorrhagic stroke must be established because treatment for the two differs completely. CT, essentially a three- dimensional X-ray, is the preferred test. It is quick, relatively inexpensive, and widely available, even in small hospitals or community clinics. Moreover, it is very sensitive for detecting intracranial bleeding because iron in the blood’s hemoglobin readily absorbs X-rays. Examples of CT scans from two patients, one with an ischemic stroke and one with a hemorrhagic stroke, are illustrated in Figure 2.
FIGURE 1 Stroke diagnosis. Upon admission to a hospital, a physician uses this decision tree to establish the most likely diagnosis and aid in subsequent treatment. First, a focal neurological deficit must be established. If it resolves spontaneously, it suggests a transient ischemic attack (TIA). If the deficit persists for more than 24 h, a stroke is suspected. The imaging results determine whether the infarct is hemorrhagic, with evidence of blood that results in high-density areas on the computed tomographic (CT) scan. If this is not the case, an ischemic infarct is the most likely diagnosis.
(CT) scans from two patients. The left image illustrates a hemorrhagic stroke in the basal ganglia; the increased signal (hyperdensity, arrow) signifies bleeding. The right is a characteristic ischemic stroke with reduced density in the infarct region (hypodensity, arrows) suggestive of stroke-associated edema. Images were kindly provided by Dr. Surjith Vattoth, Radiology, University of Alabama Birmingham.
The ischemic lesion contains more water because of edema and therefore presents with reduced density on CT, whereas the absorption of X-rays by blood creates a hyperdense image on CT in a hemorrhagic stroke.3 If bleeding is detected, any attempt to stop blood entering the brain (including surgical, if possible) must be considered. In the absence of bleeding, restoring blood flow to the affected brain region as quickly and effectively as possible is imperative. Major advances to restore blood flow using chemical or mechanical recanalization of occluded blood vessels have been made and are extensively discussed in Section 5. If the symptoms resolve spontaneously and quickly, within less than 24 h, we typically consider the insult a transient ischemic attack (TIA) as opposed to a stroke. However, the distinction between TIA and stroke is less important regarding treatment decisions because we do not have the luxury to wait 24 h before providing treatment. If, as is often the case, symptoms resolve within minutes to an hour, the diagnosis of TIA is an important risk factor for the patient, who has an elevated risk of developing a stroke in the future (5% within 1 year). It is possible to misdiagnose a stroke in an emergency room setting, where time is of the essence and diagnoses must be made quickly. A number of conditions can mimic stroke symptoms, including migraine headaches, hypoglycemia (particularly in diabetic patients), seizures, and toxic-metabolic disturbances caused by drug use. Some of these can be ruled out by simple laboratory tests; hypoglycemia is a good example. Others can be excluded through a detailed patient history and, in particular, the ability of the physician to establish a definitive history of focal neurological symptoms, ideally corroborated by an eye witness. Rapid diagnosis is facilitated by the use of a simple, 15item stroke assessment scale established by the National
I. STATIC NERVOUS SYSTEM DISEASES
3 Clinical presentation/diagnosis/epidemiology
Institutes of Health. This assessment, called the National Institutes of Health Stroke Scale, assesses level of consciousness, ocular motility, facial and limb strength, sensory function, coordination, speech, and attention.4 The pathophysiology of stroke is well understood, and the treatments available to date are effective for many patients. Unfortunately, the underlying disease causes, including atherosclerosis and hypertension, can rarely be completely removed, although a combination of lifestyle changes and chronic medical management of risk factors can reduce the likelihood of recurrence. Numerous risk factors have been identified, many of which are modifiable through changes in lifestyle or medication. By far the largest risk factor is a person’s age, which increases incidence almost exponentially, doubling with every 5 years of life. Put in perspective, only 1:10,000 persons are at risk of suffering a stroke at age 45; that number climbs to 1 in 100 by age 75. The second-leading risk factor is hypertension, which in creases stroke risk about 5-fold, followed by heart disease (3-fold), diabetes (2- to 3-fold), smoking (1.5- to 2-fold), and drug use (1- to 4-fold). Note that these risk factors are additive, and thus a 75-year-old diabetic smoker who drinks and has heart disease has a greatly compounded risk. African Americans are twice as likely to suffer a stroke as Caucasians. Men are slightly more likely than women to suffer a stroke and die from it.
7
Epidemiological data established what is often called the “stroke belt,” namely, a geographic region within the United States where annual stroke deaths are highest (Figure 3). This is readily explained by the confluence of risk factors of race, diabetes, and obesity among the population in the southern and southeastern United States. Stroke is the most common neurological disorder in the United States affecting close to 800,000 people each year. It is the fifth-leading cause of death, with ~ 150,000 stroke-related deaths annually. Many patients survive but remain permanently disabled, making stroke the leading cause of permanent disability in the United States with ~ 6.4 million stroke survivors. Globally, stroke is the second-leading cause of death with an estimated 13.7 million new stroke cases causing 5.5 million deaths. It is also a leading cause of disability with 80 million stroke survivors worldwide (GDB stroke collaborators, Lancet Neurology 2019). High-income countries including the United States and Europe have seen declining rates of incidence and mortality, most likely as a result of improved awareness and management of risk factors such as hypertension and hypercholesterolemia. However, the global burden of stroke continues to increase. Significant differences in stroke disease burden exist between countries. For example, incidence is four times higher in China than in Latin America, yet this d ifference is entirely attributable
FIGURE 3 Color-coded annual stroke deaths by region show an elevated incidence in the Southern United States, a region often dubbed the “stroke belt.” Stroke death rate for adults over 35 and including all races and all genders for the time period 2008–2010. Produced with data from Centers for Disease Control and Prevention, Atlanta, GA, USA.
I. STATIC NERVOUS SYSTEM DISEASES
8
1. Cerebrovascular Infarct: Stroke
to the average age of the population as age-adjusted stroke incidence is identical around the globe.
4 DISEASE MECHANISM/CAUSE/BASIC SCIENCE Stroke is conceptually a simple disease wherein the brain’s “plumbing” is defective. We have a fairly good understanding of causes and remedies. In its most elementary form, a stroke is the direct result of inadequate blood flow to a region of the brain, with ensuing death of neurons as a consequence of energy loss. To fully appreciate the vulnerability of the brain to transient or permanent loss of blood flow, it is important to discuss the unique energy requirements of the brain and the cerebral vasculature that delivers this energy. The brain is the organ that uses the largest amount of energy in our body. At only 2% of body mass, an adult brain uses 20% of total energy, whereas a child’s brain uses as much as 40%. The cellular energy unit is adenosine triphosphate (ATP), the majority of which is produced by the oxidative metabolism of glucose to carbon dioxide (CO2) and water. To supply sufficient ATP, an adult brain requires 150 g glucose and 72 L of oxygen per day. While the developing brain still uses a significant amount of energy for the biosynthesis of cellular constituents, particularly myelin, the adult brain has very little synthetic activity because few cells and membranes are ever replaced. Thus, the vast majority of energy is used to shuttle ions across the cell membrane to establish and maintain ionic gradients necessary for electrical signaling (Figure 4).
FIGURE 4 Cellular energy use of neurons in the brain. Adenosine triphosphate (ATP) produced in the mitochondria directly fuels ATPdriven pumps such as the Na+ K+ ATPase and the Ca2 +-ATPases and indirectly provides the energy for Na+-coupled transporters.
Of greatest importance is the extrusion of Na+ and the import of K+ through Na+/K+ ATPase. This pump not only establishes the inward gradient for Na+ needed to generate an electrical impulse or action potential but also maintains a negative resting membrane potential that neurons assume between action potentials. Moreover, the electrochemical gradient for Na+ is harnessed to transport glucose and amino acids across the membrane and to regulate intracellular pH. Therefore, these transport systems are indirectly coupled to the ATP used by the Na+/K+ ATPase. Additional important consumers of cellular ATP are Ca2 +-ATPases that transport Ca2 + against a steep concentration gradient either out of the cell or into organelles. Intracellular Ca2 + is maintained around 100 nM, which is 10,000-fold lower than the 1 mM concentration of Ca2 + in the extracellular space. Ca2 + functions as a second messenger in only a very narrow concentration range of 100– 1000 nM and therefore must be carefully regulated by the Ca2 +-ATPases. Any increase above this range activates enzymes and signaling cascades that are largely destructive (discussed in more detail later in this chapter). Finally, ATP serves as an important source for high-energy phosphates that can attach to proteins and enzymes through phosphatases that act as on/off switches to regulate the activity of these proteins and enzymes. ATP stores in neurons are exhausted after only 120 s. Therefore, neurons must continuously produce ATP from glucose via oxidative metabolism of glucose in the mitochondria. Glucose is the most readily available energy source throughout the body, and most cells can store some readily available glucose in the form of glycogen granules. These glycogen granules are a polysaccharide of glucose that can be quickly converted back to glucose when needed for energy. Unfortunately, neurons do not contain glycogen stores and therefore rely on a constant, uninterrupted supply of glucose from the blood. From an evolutionary point of view, the cellular space saved by giving up energy stores allows the important benefit of an increased number of nerve cells packed into a finite cranial space. To meet its high-energy demands, it is also essential that the brain metabolizes all glucose in the most effective way possible, by oxidative metabolism, which yields 36 mol ATP/mol glucose. This far exceeds the anaerobic glycolytic production of ATP, which only yields 2 mol ATP/mol glucose. Unlike most other cells in the body, neurons are not able to switch to glycolysis in the absence of oxygen, necessitating a constant delivery of sufficient oxygen. The convergence of high-energy demand, the absence of glycogen stores, and exclusively aerobic metabolism makes the brain uniquely vulnerable to injury in situations where glucose or oxygen supply is disrupted. Rare conditions limit only one of these substrates. For example, hypoglycemia may occur in a diabetic patient who receives an excess amount of insulin, and anoxia can occur in a patient in a near-drowning
I. STATIC NERVOUS SYSTEM DISEASES
4 Disease mechanism/cause/basic science
situation who stops breathing. In general, however, cerebrovascular infarction is the result of reduced blood flow that limits both glucose and oxygen delivery; this condition is called ischemia. It is important to note that while energy consumption throughout the body varies with activity, most notably in skeletal muscles and the heart, the brain’s metabolic activity is fairly constant and not measurably affected by changes in mental state. Therefore, the overall regulation of blood flow to the brain simply ensures a constant flow of oxygenated blood to the brain. However, regional differences in energy consumption occur and give rise to the blood oxygen concentrations measured by functional magnetic resonance imaging. For every region with enhanced regional blood flow, there is another that has reduced blood flow, effectively canceling each other for a constant metabolic activity. The cerebral vasculature receives its main supply of oxygenated blood via the two common carotid arteries on each side of the neck, which branch into the internal carotid artery (ICA) and external carotid artery (ECA), respectively (Figure 5). The ICA is the predominant supply line, carrying ∼ 75% of the total blood volume to the brain, whereas the ECA primarily feeds the neck and face. Two vertebral arteries at the back of the neck provide an additional mi-
FIGURE 5 The cerebral vasculature in a schematic view. The main supply of oxygenated blood to the brain is through the two common carotid arteries on each side of the neck, which branch into the internal (ICA) and external carotid arteries (ECA), respectively. The ICA is the predominant supply line, carrying ~ 75% of the total blood volume to the brain, while the ECA primarily feeds the neck and face. The ICA ends by dividing into the middle (MCA) and anterior cerebral (ACA) arteries. Two vertebral arteries at the back of the neck provide an additional minor supply pathway for the brain; this pathway becomes important in situations where the carotids are narrowed or blocked. Image by Ian Kimbrough, PhD.
9
FIGURE 6 Perfusion fields of the major cerebral arteries. ACA, anterior cerebral artery; MCA, middle cerebral artery; PCA, posterior cerebral artery.
nor supply pathway for the brain; this pathway becomes important in situations where the carotids are narrowed or blocked. The ICA ends by dividing into the middle cerebral artery (MCA) and anterior cerebral artery (ACA). The MCA is the largest branch and divides into 12 smaller branches; together, these 12 branches supply almost the entire cortical surface, including the frontal, parietal, temporal, and occipital lobes (Figure 6). At its stem, the MCA gives rise to additional vessels that supply the midbrain, including the globus pallidus and caudate nucleus. The ACA supplies primarily the frontal lobes. The vertebral arteries supply the cerebellum and medulla. The basilar artery branches off the vertebral artery and supplies the pons and lower portions of the midbrain, hypothalamus, and thalamus. The posterior cerebral artery branches off the basilar artery and feeds the occipital lobe. Many of the major vascular branches are interconnected and form a network that allows blood to circumvent obstructions if present. One particular structure that deserves mentioning is the circle of Willis. This ringlike connection of the cerebral vasculature is established by the anterior commissure connecting the left and right branches of the ACAs and the posterior commissures connecting the posterior cerebral arteries. By contrast, smaller arteries less than 100 μm in diameter are end arteries that are not interconnected, and any blockage results in loss of perfusion to the innervated brain region. Note that from an evolutionary point of view, the brain’s vasculature is a catastrophe waiting to happen. With the massive expansion of the cortex in humans, the MCA has to supply a much larger part of the brain than originally intended. Importantly, almost all of the brain supplied by the MCA is considered our “eloquent brain,” areas responsible for motor and sensory function, language, and cognition. Hence, a reduction in blood flow through the MCA causes a disproportionate functional loss of eloquent brain function in humans.
I. STATIC NERVOUS SYSTEM DISEASES
10
1. Cerebrovascular Infarct: Stroke
FIGURE 7 Vascular cast of a human brain shows the extensive branching of vessels into finer and finer structures. The cast was prepared by injection of a plastic emulsion into the brain vessels, and, upon hardening, the brain parenchymal tissue was enzymatically dissolved. Reproduced with permission from Ref. 5.
Each heartbeat delivers about 70 mL of oxygenated blood to the aorta, 10–15 mL of which are allocated to the brain. Every minute about 500 mL of blood circulate through the brain. To ensure constant perfusion, pressure and blood flow are highly regulated. The first line of regulation is via the arterial walls of the major arteries, which constrict in response to increases in blood pressure. Arterioles are exquisitely sensitive to changes in the partial pressure of CO2 such that when the CO2 content increases, indicating high metabolic activity, arterioles dilate. This dilation causes increased blood flow and enhances delivery of oxygenated blood. When CO2 decreases, vessels constrict to reduce blood flow. As noted earlier, functional activity within subregions of the brain adjusts regional blood flow without affecting the overall delivery of blood to the brain, which remains about 500 mL/min. The brain is very sensitive to even a modest decrease in blood flow. A drop from a normal value of 20–30 to 16–18 mL/100 g tissue causes infarction within 1 h, and any further reduction kills brain tissue in just minutes. As illustrated in a vascular cast of a human brain (Figure 7), arterioles branch extensively, giving rise to capillaries so small that erythrocytes have to bend to fit through their lumen. This site is where the major exchange of glucose and blood gases with brain tissue occurs. Since diffusion of gas in tissue is very limited, capillaries reach within about 50 μm of any neuron throughout the brain.
for survival. Coagulation involves two steps: the initial formation of a cellular plug by circulating platelets and a secondary reinforcement of this plug by an aggregate of fibrin fiber strands. The end product is a blood clot or thrombus that seals the vessel wall. The clotting process is well understood and involves numerous clotting factors. Clotting proceeds along two pathways: the intrinsic (contact activation) pathway and the extrinsic (tissue factor) pathway. The latter is the pathway we are most concerned within the context of stroke. Here, following damage to the vessel wall, the serine protease thrombin causes the production of fibrin from fibrinogen. Fibrin fibers then form the hemostatic plug. Normally, as a wound heals, blood clots are dissolved through a process called fibrinolysis. The main protein mediating the dissolution of the plug is plasmin. Plasmin is derived from a precursor molecule, plasminogen (PLG), which is a component of blood serum produced in the liver. The cleavage of the precursor PLG to the active plasmin is catalyzed by tPA, a serine protease produced by endothelial cells in blood vessels (Figure 8). Plasmin in turn helps degrade the clot. Therefore, tPA is an important regulator of clotting, and recombinant tPA is an effective thrombolytic agent, often called a “clot-buster.” We discuss its clinical use later in this chapter. In many stroke patients, the obstruction of blood flow is the result of atherosclerosis, a chronic vascular disease that causes a thickening of the arterial wall with plaque deposits that narrow the vessel lumen (Figure 9). Plaques contain Ca2 + and fatty acids, including triglycerides and cholesterol, as well as macrophages and white blood cells. Atherosclerotic plaque formation is promoted by low-density lipoproteins, often called “bad cholesterol,” which enters into the vessel wall and causes an immune response. This in turn sends macrophages and T-lymphocytes to the affected vessel, causing them to aggregate locally. In addition to a narrowing of the arterial lumen, this causes a hardening of the vessel wall and inflammation of the surrounding smooth muscle.
4.1 Causes of Vessel Occlusions: the Thrombolytic Cascade The loss of blood flow in stroke patients is the result of insufficient arterial flow due to vessel narrowing, complete blockage, or rupture. The underlying causes are directly related to biochemical events involved in the formation of blood clots. The ability of blood to coagulate and stop bleeding from wounds (hemostasis) is essential
FIGURE 8 Breakdown of thrombotic deposits or blood clots by tissue plasminogen activator (tPA). Schematically shown is the lumen of an arterial blood vessel formed by capillary endothelial cells. tPA activates the conversion of plasminogen (PLG) to plasmin, which catalyzes the breakdown of fibrin clots into fibrin degradation products. Plasmin can be inhibited by antiplasmin (AP).
I. STATIC NERVOUS SYSTEM DISEASES
4 Disease mechanism/cause/basic science
11
FIGURE 10 Schematic view of thromboembolic ischemia showing a reduction of blood flow by the buildup of thrombolytic plaque material on the inside of the vessel wall. This thrombus can break off and fragment; so-called emboli travel with the blood and can plug finer vessels. Occlusions therefore often have been called thromboembolic because no distinction is made between stationary buildup and floating debris. Drawing by Emily Thompson, PhD.
FIGURE 9 Examples of atherosclerosis critically narrowing vessel lumen and thereby significantly reducing blood flow. From Wiki-Commons.
Atherosclerosis can affect essentially all arterial vessels in the body and is considered a chronic condition that typically proceeds asymptomatically for many years or decades. Even a small narrowing of the vessel lumen has a profound effect on blood flow and hence energy delivery to the brain. The weakened vessel wall becomes prone to rupture and may cause a hemorrhagic stroke, particularly in patients with chronic hypertension.. Plaques are covered by a fibrous cap of collagen, which is highly clot promoting. This cap is typically weak and prone to rupture. If it enters the bloodstream, it forms a thrombus that attracts additional platelets and white blood cells. It eventually attaches at a vessel branch point, where it occludes blood flow. A small particle breaking off from a thrombus or the plaque is called an embolus. It floats along with the blood into smaller and smaller penetrating arteries where it eventually becomes lodged. This process is illustrated in Figure 10. We refer to ischemic strokes caused by thrombi or emboli as thromboembolic strokes. Historically, we assumed that atherosclerosis is mostly influenced by diet, with fatty and highly processed foods being often considered the major culprits. Hence, modern humans would be predisposed to stroke as a consequence of their diet. However, recent
use of MRI to examine blood vessels in mummified bodies of four different ancient human populations showed the presence of atherosclerotic plaque comparable to that of modern humans even at a relatively young age.6 Plaques were also present in mummies suspected to have consumed a meat-free diet. These findings challenge the commonly held notion that atherosclerosis can be readily explained by dietary choice. They do not, however, question the obvious link between atherosclerosis and stroke.
4.2 The Ischemic Cascade The loss of blood flow resulting from either leakage or blockage of a blood vessel starts what is known as the ischemic cascade. From decades of experimentation in animals and humans, we have an excellent understanding of the cellular and molecular changes that occur in the affected brain. The time course and the major cellular changes are illustrated graphically in Figures 11 and 12. Following the fairly rapid exhaustion of ATP, the neuronal membrane slowly depolarizes as the Na+/K+ ATPase fails to maintain the resting membrane potential. After just a few minutes, most affected neurons reach the activation threshold for Na+ channel-mediated action potentials, causing a transient period of hyperexcitability that further depolarizes the membrane. The depolarization opens voltage-gated Ca2 + channels and causes the removal of the Mg2 + block that normally safeguards the N-methyl-d-aspartate (NMDA) receptor. This in turn causes further uncontrolled influx of Ca2 +. Note that while NMDA receptors typically cause a brief excitatory postsynaptic current that is largely due to Na+ influx, the channel per se has fivefold higher permeability to Ca2 + than Na+. At this stage, all pathways for Ca2 + entry are
I. STATIC NERVOUS SYSTEM DISEASES
FIGURE 11 Time course of ischemia; cellular and molecular events that contribute to injury are indicated. Different pathways lead to either apoptotic or necrotic cell death.
FIGURE 12 The ischemic cascade and pathways activated as a result. Initiated by energy failure, (1) presynaptic release of glutamate aber-
rantly activates N-methyl-d-aspartate receptors (NMDA-Rs). (2) The uncontrolled influx of Ca2 + and Na+ as a result of NMDA-R activation causes (3) cell swelling, (4) activation of destructive enzymes, (5) mitochondrial damage, and the generation of reactive oxygen species (ROS). These processes culminate in (6) DNA damage and (7) membrane damage during apoptosis. This may cause (8) microglial activation and (9) leukocyte recruitment from the peripheral blood. Drawing by Emily Thompson, PhD.
4 Disease mechanism/cause/basic science
open and the Ca2 +-ATPase that normally sequesters Ca2 + into organelles is inactivated by the loss of ATP. To aggravate matters further, surrounding astrocytes also exhaust their small glycogen stores and their failure to produce ATP results in astrocytic depolarization. This dissipates the transmembrane gradient for Na+ that is required for uptake of glutamate (Glu) into astrocytes. The depolarized neurons release more Glu, which is no longer cleared by astrocytes. This causes a catastrophic scenario whereby sustained activation of neuronal Glu receptors causes Ca2 + influx, which feeds forward on further vesicle fusion and Glu release. Ultimately, within 5–10 min, irreversible neuronal death begins in the core of the ischemic lesion. Early neuronal and glial cell death occurs primarily through a necrotic pathway, where the influx of ions, including Na+ and Cl−, causes cytotoxic edema or cell swelling as water follows these ions into the cells. This in turn ruptures cell membranes, causing spillage of cytoplasm and creating a toxic extracellular space. Later, the Ca2 + increase activates a secondary programmed cell death cascade, which is explained further in the following.
13
FIGURE 13 The ischemic penumbra is the area surrounding a
4.3 The Ischemic Penumbra
stroke lesion where blood flow is still reversible and where tissue can potentially be rescued if blood flow is restored in a timely manner. It can be divided into two regions: one in which diffusion is abnormal, suggesting structural damage, and one in which diffusion is normal but perfusion of tissue with blood is abnormal. This area in particular can be rescued through rapid restoration of blood flow. Drawing by Emily Thompson, PhD.
The ischemic lesion is surrounded by brain tissue that still receives some blood flow via collaterals; however, this blood flow is not of sufficient quantity to maintain normal brain function. This hypoperfused brain tissue, schematically illustrated in Figure 13, is called the ischemic penumbra and is the focus of all intervention strategies. Here, neurons and glial cells are in a latent state, paralyzed from participating in proper neuronal function but clinging to life. If perfusion occurs within a reasonable amount of time (6–24 h), the ischemic penumbra can recover completely. If not, tissue in the penumbra gradually dies and becomes part of the expanding ischemic core. The viability of the ischemic penumbra has been extensively studied in rodents and monkeys, and our best estimates suggest that neurons in the penumbra remain viable for at least 3–6 h, although some studies suggest that this time period may be as long as 24–48 h. However, based on consensus data from many studies, 4.5 h has been adopted as the critical time period during which reperfusion provides maximal benefit, and this is often called the “window of opportunity.” In agreement with this suggestion, chemical recanalization using tPA, discussed in the following paragraph, provides maximal benefit to patients who are only less than 4.5 h removed from the insult, with declining benefits between 4.5 and 6 h and essentially none thereafter. Realizing this window of opportunity, clinicians are urging patients to seek rapid attention because “time lost is brain lost.” Our understanding of the ischemic penumbra is not quite as clear as that of the ischemic core. It seems, however, that neuronal death in the penumbra p rimarily
occurs via programmed cell death, called apoptosis. Increases in Ca2 + to supraphysiological, but not catastrophic, levels activate several of the apoptosis- promoting pathways. Activation of these pathways then culminates in the release and activation of caspases, a family of cysteine proteases that cleave cellular proteins, resulting in slow cellular disassembly. This goes hand in hand with changes to the mitochondrial membrane, which releases proapoptotic molecules such as cytochrome C. The opening of a large ion channel, called the mitochondrial transition pore, breaks down the voltage gradient across the mitochondrial membrane required for the production of ATP. This in turn further compromises the production of energy by brain tissue, further contributing to a vicious cycle of neuronal cell death. The ischemic penumbra is also a site of extensive inflammation. Both astrocytes and microglial cells are activated during a stroke, and blood-borne immune cells, including leukocytes, T-lymphocytes, and natural killer cells, infiltrate the region as well. These inflammatory cells release cytokines, which in turn exacerbate the neural injury by stimulating nitric oxide production and enhancing NMDA-mediated excitotoxicity. In both the ischemic core and the penumbra, the integrity of the blood-brain barrier is compromised. This is partly due to the activation of matrix metalloproteinases and the retraction of astrocytic endfeet. The breakdown of the blood-brain barrier enhances penetration of blood-borne toxins and immune cells, each of which contributes to the further decline of tissue health.
I. STATIC NERVOUS SYSTEM DISEASES
14
1. Cerebrovascular Infarct: Stroke
It is commonly assumed that the ischemic core cannot be rescued, whereas the penumbra can, hence the objective of treatment is to maximize survival of cells in the ischemic penumbra by limiting excitotoxic or inflammatory injury and to restore normal blood flow. Therefore, noninvasively defining whether a penumbra exists for a given patient and determining the size of this area is of great importance. A combination of imaging techniques now allows us to gain this insight. Specifically, magnetic resonance imaging studies can readily distinguish between perfusion and diffusion within tissue. Diffusion-weighted images show structural, and hence permanent, destruction, thereby identifying the ischemic core. Perfusion images, on the other hand, show changes in relative perfusion or blood flow and presumably the brain region with abnormally low, but not absent, blood flow. By overlaying these two images, one can identify an area in which there is a mismatch between diffusion and perfusion and this area represents the penumbra. Example images in Figure 14 illustrate this approach. While a potentially powerful predictor of vulnerable brain tissue and proven to be of clinical value, this approach requires imaging capacity only available at specialized hospitals, and sufficient time to acquire images without compromising rapid treatment. Decades of research using animal models of stroke have provided a fairly detailed understanding of the cellular and molecular changes that cause necrotic or apoptotic cells death, and example is illustrated in Figure 12. Many of the identified molecules and pathways predicted cures, yet, as further discussed in the
FIGURE 14 Mismatch of perfusion and diffusion using magnetic resonance imaging can identify the presence and size of the penumbra. Red demarcates the infarct zone, with irreversible injury, blue demarcates the zone with reduced blood flow containing at-risk tissue, and the green mismatch zone shows the area in which at-risk tissue can be saved. Reproduced with permission from Ref. 7.
following, none of the treatments that were successful in animal models of stroke translated into effective treatments in humans. The molecule that received the most attention is the NMDA receptor, believed to be the pivotal pathway for the entry of toxic Ca2 + in ischemia (Figure 15).
FIGURE 15 N-Methyl-d-aspartate receptor (NMDA-R) as mediator of excitotoxicity and target for therapeutic drugs. Permeation of Ca2 + via NMDA-Rs (left) causes activation of the mitogen-activated protein kinase (MAPK) cascade as well as release of cytochrome c from the mitochondria, each culminating in apoptotic cell death. The NMDA-R harbors many regulatory sites that can be exploited for therapy (right), including sites for the drugs memantine and MK-801 as well as the coagonist sites for Zn2 + and glycine (Gly). Reproduced with permission from Ref. 8.
I. STATIC NERVOUS SYSTEM DISEASES
4 Disease mechanism/cause/basic science
4.4 The NMDA Receptor and Glutamate Excitotoxicity The NMDA receptor (NMDA-R) is a Glu-gated cation channel that is a member of the ionotropic Glu receptor family. Like the other family members, it is a heteromeric channel composed of four subunits, namely, two GluN1 and two GluN2 subunits. The GluN2 subunit comes in four isoforms named GluRN2A–D. The channel is gated by the binding of Glu, but it requires a modulatory site to be occupied by glycine. The channel distinguishes itself from all other Glu receptors by its multiple modulatory sites. In addition to the glycine site, which can bind d-serine instead of glycine, there are sites for modulation via nitric oxide, H+, and phencyclidine. The channel is the target of several anesthetic drugs, including ketamine and phencyclidine (“Angel Dust”), which are often used as recreational hallucinogenic drugs. Other common agonists of differing potency include ethanol, memantine, dextrorphan, and methadone. The feature that is of greatest relevance for stroke is the unique gating of the channel by voltage and intracellular Mg2 +. Normal, fast, excitatory synaptic transmission occurs via the α-amino-3-hydroxy5-methyl-4- isoxazolepropionic acid (AMPA) and kainate-type Glu receptors. These too are cation- permeable, ionotropic Glu receptors that open in response to Glu. NMDA-Rs, in contrast to AMPA-Rs, are normally not available for activation by Glu because, at the resting potential of the postsynaptic membrane, a Mg2 + ion is lodged in the pore of the channel, preventing it from allowing ions to permeate. This Mg2 + binding is dependent on voltage, and with depolarization of the membrane, Mg2 + is released from the pore. Under normal physiological conditions, Mg2 + keeps the NMDA-R closed at all times. Should the postsynaptic terminal be depolarized, however, the Mg2 + ion pops out, causing cations to flux through NMDA-Rs. While the vast majority of the current is carried by Na+, which is the most abundant extracellular cation, the NMDA pore is fivefold more permeable to Ca2 + than to Na+. As a consequence, a considerable amount of Ca2 + fluxes through the NMDA-R. This Ca2 + influx plays an important role during learning and memory, allowing the NMDA-R to function as a coincidence detector. As more extensively discussed in Box 1, Chapter 4, only if multiple signals arrive on the same terminal within 50 ms of each other is the Mg2 + removed. Removal of the Mg2 + ion causes Ca2 + influx, thereby signaling the cell of the coincident occurrence of two events. This is believed to be an essential component of learning and memory. In the situation of a stroke, the coincident activation is bypassed by the chronic, long-lasting depolarization of the postsynaptic membrane due to energy failure. This energy failure permanently removes the Mg2 + block, opening the flood
15
gates for Ca2 + to enter unimpeded through this receptor. Hence, the very protein that is so critical in allowing for human learning can also make us vulnerable to cell death by excitotoxicity following a stroke. The concept of Glu excitotoxicity was originally introduced by Olney in the 1960s and has been extensively studied since. Excitotoxicity appears to be the final common death pathway in numerous disorders, including many neurodegenerative diseases. The precipitating cause of the chronic depolarization of the neuronal membrane may differ in each case, yet the principal involvement of the NMDA-R as an influx pathway for Ca2 + is shared. While NMDA-R plays the most important role in ischemic neuronal death, other family members participate or can substitute. The depolarization that occurs as a result of activation of AMPA or kainate-type Glu receptor depolarizes the postsynaptic cell, thereby allowing Ca2 + to enter via voltage-gated Ca2 + channels. Furthermore, at least one variety of the AMPA receptor is also permeable to Ca2 +.
4.5 Role of Glutamate While Olney suspected Glu as a major toxin in disease, showing its contribution to stroke in animals and humans suffering a stroke was necessary. Early studies measured Glu directly by microdialysis, a technique through which the extracellular milieu in the brain can be sampled continuously and with high accuracy. They demonstrated that occlusion of the middle cerebral artery (MCA) causes a 60- to 100-fold increase in Glu within a relatively short time. The increase extends to the ischemic penumbra and even affects the composition of the cerebrospinal fluid (CSF) bathing the brain. CSF can be examined by lumbar puncture in patients in whom microdialysis would rarely be feasible unless the patient underwent surgery. Glu in the CSF rises eightfold in stroke patients and has been suggested to have predictive value for disease severity.9 It is important to note that persistent increases in Glu, such as that measured after a stroke, require the neuroprotective uptake of Glu by astrocytes to fail, suggesting that astrocyte impairments in the penumbra contribute to the progression of disease.
4.6 NMDA Inhibitors to Treat Stroke Having identified the role of the NMDA-R in stroke, examining the many modulatory sites of this receptor and how they may be exploited therapeutically is worthwhile. In principle, blocking NMDA-R ameliorates neuronal death. However, it also impairs most higher cognitive function. This is exemplified by ketamine, used as a powerful anesthetic, or by phencyclidine, a potent, mind-altering substance, both of which have
I. STATIC NERVOUS SYSTEM DISEASES
16
1. Cerebrovascular Infarct: Stroke
been associated with cognitive impairment. The ideal drug would therefore block only those NMDA-Rs that experience a pathophysiological depolarization by catastrophic increases in Glu, while sparing those receptors exposed to physiological concentrations. Such drugs actually exist and are, generally speaking, poor NMDA antagonists. One of these, memantine, is an open-channel blocker that inhibits NMDA channels poorly under physiological conditions but becomes more effective as Glu concentrations increase. Memantine has been used in experimental clinical trials for amyotrophic lateral sclerosis and dementia. It is very effective in animal models of stroke, where it almost completely protects the brain after MCA occlusion; it is currently being evaluated in a placebo-controlled Phase 1 trial (www.clinicaltrials.gov, identifier NCT02144584). There is hope that other open-channel NMDA blockers may be developed for future use in the clinical treatment of stroke.
4.7 Effect of Temperature Most biological processes are dependent on temperature. In the brain, ion transport via the various ATPases changes two- to threefold with every 10 °C change in temperature. Because the ischemic cascade is driven by a loss of function of these ATPases, one would expect that increases in temperature would accelerate damage, whereas lowering temperature would slow the injury.
This is indeed observed in rats after MCA occlusion, where the development of an infarct can be significantly reduced by hypothermia10 (Figure 16). Interestingly, some of the early neuroprotective trials using rats and the NMDA antagonist MK-801 that led to the development of many other NMDA antagonists for use in stroke are now being reinterpreted after it was found that MK-801 lowers the brain temperature by about 2 degrees. Hence, the neuroprotective effect seen with MK-801 may have been partially attributable to the hypothermia rather than just the blockage of NMDA-Rs. A correlation with fever and poor stroke outcome was reported as early as the 18th century,10 and clinical studies now conclusively show that even slightly elevated temperature after an acute stroke leads to a much worse outcome and is associated with significantly higher risk of death.12 From a mechanistic point of view, it is interesting that clinical studies have also shown a twofold increase in Glu in the CSF of ischemic stroke patients who had elevated body temperature. Moreover, clinical studies of stroke patients showed that a stroke per se can cause a local increase in brain temperature, suggesting active brain inflammation in the ischemic penumbra. These findings provide ample reasons to consider lowering brain temperature in patients with a stroke to improve outcome as further discussed. At a minimum, one must control fever when present in stroke patients using drugs such as acetaminophen or ibuprofen.
FIGURE 16 Effect of hypothermia on stroke volume after middle cerebral arterial occlusion in a rat. Reduction in temperature to 32 °C greatly reduces the size of the ischemic lesion with no further benefit when reduced to 27 °C. Tissue damage is indicated by more intense colors. Normothermic (37 °C) animals show a large consistent cortical and subcortical infarct at multiple coronal levels. Both intraischemic and postischemic hypothermia to 32 °C are associated with marked reductions in the frequency of cortical infarction. intra, intraischemic; post, postischemic. Reproduced with permission from Ref. 11.
I. STATIC NERVOUS SYSTEM DISEASES
5 Treatment/standard of care/clinical management
4.8 Stroke Genetics Stroke is not a genetic disease in the strictest sense. We do not yet have evidence for single or multiple genes that, when inherited, are sufficient to cause stroke, as is the case in Alzheimer or Parkinson Disease. However, a number of individual stroke risk factors are heritable, including hypercholesterolemia, atherosclerosis, diabetes, heart disease, and therefore can “run in families.” It is therefore reasonable to expect that conglomerations of genes can be identified that together increase the risk of stroke. A genome-wide search of human stroke patients suggests that up to 38% of stroke cases show some heritability.13 However, this information does not yet allow us to determine a risk factor based on genetic analysis for any individual. Note that the field of complex genetics is still in its infancy as only recently large patient datasets became available along with the computing power to study genome-wide trait associations in such large patient cohorts. Hence, there is hope that ultimately a genetic test may be developed that will compute a relative stroke risk and may even identify the stroke subtype of greatest risk.
5 TREATMENT/STANDARD OF CARE/ CLINICAL MANAGEMENT From the previous discussion on mechanisms underlying stroke injury, it is clear that the greatest therapeutic benefit can be achieved by rapid reperfusion of the infarcted brain, particularly to rescue the ischemic penumbra that is at high risk of tissue destruction. In principle, there are two approaches to removing the obstructive embolus or thrombus: (1) chemical dissolution using tPA, or (2) mechanical retrieval using endovascular thrombectomy with a wire-guided retrieval device. In either case, time is of the essence, and immediate admission to the hospital is required to achieve a maximum therapeutic benefit. Once admitted to the emergency room, the clinical team must work expeditiously to determine the course of action deemed most beneficial. During the initial evaluation, the decision tree in Figure 1 is followed to categorize the underlying event as either ischemic or hemorrhagic as the latter excludes both mechanical and chemical recanalization. This is best accomplished in a specialized Stroke Center that has optimized the work flow. Not surprisingly, stroke centers often achieve better clinical outcomes.
5.1 Chemical Thrombolysis Using Intravenous tPA The only US Food and Drug Administration-approved drug for chemically opening vessels is the recombinant human plasminogen activator (alteplase/activase). Its use is now considered standard of care provided certain
17
inclusion criteria are met.14 Most important of these is the time that has elapsed since the stroke. Typically, treatment with tPA must begin within 4.5 h from the beginning of stroke symptoms. If the time of the event cannot be conclusively determined, tPA treatment cannot be given. Unfortunately, this is often the case particularly if the stroke occurred at night and the victim is not found until the next morning. If no one witnessed the attack, the last time the patient was seen in a normal condition will be used as a substitute. Once in the hospital, a number of additional tests must be completed for a patient to be eligible all within this 4.5-h time window. These include a noncontrast CT scan of the head to rule out hemorrhage and a blood glucose test through finger-stick to rule out hypoglycemia. Other factors, such as recent surgery, a history of head trauma or stroke, or high blood pressure (> 185 mmHg systolic), would rule out tPA use as well. As a result of these stringent criteria, fewer than 5% of stroke patients are eligible to receive tPA. Following the outcome of two large phase III clinical studies conducted by the National Institute of Neurological Disorders and Stroke,15 tPA is administered at 0.9 mg/kg, with 10% injected as a single bolus and the remaining dose administered through continuous infusion over the next 60 min. During this infusion period, the patient is monitored for any signs of intracerebral hemorrhage, determined by neurological signs such as headache, nausea, or vomiting. These signs would be cause for immediate termination of tPA infusion. The largest risk associated with tPA is intracranial hemorrhage, which occurs in 6–7% of cases and carries a very high mortality of 45%.15 Data reported in the original clinical study suggest that patients treated with tPA within 3 h were at least 30% more likely to have minimal or no disability 3 months after treatment. Additional studies have since explored the use of tPA for up to 4.5 h after the incident and report smaller but still significant improvements of outcome with an acceptable increase in risk. Based on these findings tPA is now approved for up to 4.5h after an infarct.16 A fascinating vignette from the original tPA study is the “Lazarus effect” (named after the miraculous resurrection of Lazarus 4 days after his death). Of patients treated with tPA, 20% showed a dramatic, indeed almost miraculous, improvement 24 h after treatment. Such a phenomenon has been previously reported following cardiac arrest, where spontaneous, unexplained reperfusion of the heart occurred in some patients after resuscitation efforts were suspended. A newer thrombolytic agent with similar mechanism of action, Tenecteplase, was recently introduced as alternative to Alteplase. It has the advantage of being administered as a single injected dose. In clinical studies, it was found to be equally effective and showed a similar safety profile and hence may be a more convenient alternative.
I. STATIC NERVOUS SYSTEM DISEASES
18
1. Cerebrovascular Infarct: Stroke
5.2 Mechanical Thrombolysis by Endovascular Therapy for Ischemic Stroke Microcatheters have long been used to place stents into cardiac blood vessels. These devices are small enough to be inserted into cerebral arteries as well and have been refined into sophisticated clot dissolution and retrieval devices. In their simplest form, repeated passage of a wire through a thrombus helps to disintegrate it mechanically. In a more advanced form, these devices contain a soft silicon balloon that can be expanded to open a collapsed artery or a wire mesh to retrieve an embolus and restore blood flow. This approach is particularly promising for opening larger arteries such as the MCA or ICA. Successful opening of an artery and placement of a stent is shown in Figure 17, with Figure 18 illustrating balloon angioplasty and stent placement. In 2015, five large, multicenter clinical studies conducted in different parts of the world independently explored whether the combined use of mechanical and chemical thrombectomy may be superior to tPA treatment alone. All used a wire-guided retrieval device called a stent retriever. It is threaded through an artery in the groin up to the blockage where the stent opens and retrieves the clot. The five studies differed slightly regarding the relative timing of tPA versus mechanical retrieval. All five studies reported that combined chemical and mechanical thrombectomy is far superior than tPA alone.18 Up to 100% brain reperfusion was achieved 24 h after combined treatment, and 60% of patients were
functionally independent 90 days after treatment compared to 35% that received only tPA. Not surprisingly, the greatest benefits were achieved in patients in which brain images suggested the presence of a salvageable penumbra. Indeed, in patients in which the clinical deficit was larger than would be expected based on imaging data, mechanical thrombectomy administered up to 24 h after the stroke, still achieved functional independence in 49% of patients compared to 13% receiving standard of care.19 Taken together, these clinical studies
FIGURE 18 Schematic depiction of mechanical recanalization using balloon angioplasty with placement of a stent. Illustration by Emily Thompson, PhD.
FIGURE 17 Successful recanalization of the anterior cerebral artery (ACA) and middle carotid artery (MCA) by balloon angioplasty: (left) before and (right) after angioplasty and placement of a stent. (A) Left internal carotid artery cerebral angiography demonstrates severe stenoses of the ACA and MCA, which are confluent at the carotid terminus. (B) After angioplasty alone of the ACA and stent-assisted angioplasty of the MCA, there is normal caliber of the previously stenosed ACA and MCA. Arrows denote the proximal and distal ends of the stent. Reproduced with permission from Ref. 17.
I. STATIC NERVOUS SYSTEM DISEASES
5 Treatment/standard of care/clinical management
nequivocally established that aggressive mechanical u removal of blockages, ideally combined with chemical thrombolysis, yields significant improved patient outcomes particularly when brain imaging can establish the presence of viable tissue in the penumbra. Combined chemical and mechanical thrombectomy is now considered standard of care in patients that meet the inclusion criteria who have large vessel occlusion to the anterior brain circulation. Unfortunately, this aggressive approach requires specialized stroke centers that have the necessary experience, imaging equipment, and surgical skills to rapidly administer endovascular thrombectomy. These capabilities are still limited even in many industrialized countries and essentially absent in the rest of the world.
5.3 Anticlotting Factors to Prevent Recurrence Unlike acute thrombolysis with tPA, the prevention of clotting through systemic treatment with antiplatelet agents in the acute presentation of a stroke has not been shown to be effective and indeed carries an increased risk. This is not to be confused with the proven benefit of such drugs to prevent stroke recurrence, as further discussed in the following.
5.4 Treatment of Hemorrhagic Stroke Hemorrhagic strokes are very difficult to treat and carry a large (50%) mortality. Immediate treatment steps depend greatly on the condition of the patient. Acute control of systolic blood pressure to 150 De novo mutations
Febrile, generalized tonic– clonic, Dravet syndrome
Slow inactivation, broadening of the action potential, persistent current
SCN2A
Nav1.2 Na+ channel alpha subunit
Autosomal dominant
Benign familiar epilepsy syndrome
Slow inactivation, broadening of the action potential, persistent current
KCNA1
Kvl.l Shaker related K+ channel
Autosomal dominant
Focal epilepsy
KCNQ2
Kv7.2, M-current cation current
Autosomal dominant
KCNQ3
Kv7.3, M-current, cation current
Autosomal dominant
HCN1
Slowed repolarization of the action potential noninactivating currents that normally raise the threshold for Benign familiar epilepsy excitability, loss of function increases syndrome with tonic–clonic excitability seizures Benign familiar epilepsy syndrome
Noninactivating currents that normally raise the threshold for excitability, loss of function increases excitability
Hyperpolarization-activated De novo mutations K+ channel
Temporal lobe epilepsy
Shortened refractory period, enhanced excitability
KCNMA1
KCal.l, MaxiK Ca2 +-activated K+ channel
De novo mutations
Generalized epilepsy
Contributes to repolarization, stabilize resting membrane potential, loss of function causes hyperexcitability
KCNT1
KCa4.1, Ca2 +-activated K+ channel
De novo mutations
Generalized epilepsy, Stabilize resting membrane nonconvulsing malignant potential, loss of function causes migrating partial seizures hyperexcitability of infancy is a rare epileptic encephalopathy
CACNA1A
P/Q type Ca2 + channel
Autosomal dominant
Absence seizure
CACNA1H
T-type Ca2 + channels
Autosomal dominant
Childhood absence seizure Pacemaker, enhances pyramidal cell bursting
CHRNA4
nACH receptor alpha-4 subunit
Autosomal dominant
Autosomal dominant nocturnal frontal lobe epilepsy
Regulates excitability
CHRNB2
nACH receptor beta-2 subunit
Autosomal dominant
Autosomal dominant nocturnal frontal lobe epilepsy
Regulates excitability
CHRNA2
nACH receptor, alpha-2 subunit
De novo mutations
Autosomal dominant nocturnal frontal lobe epilepsy
Regulates excitability
GABRG2
GABA-A receptor, gamma-2 De novo mutations subunit
Febrile seizure, absence seizure
Decreases tonic inhibition
Required for neurotransmitter release, which, if mutated, is altered
Table established based on data provided in Lerche et al. J Physiol. 2013;591(4):753–764; Jasper’s Basic Mechanisms of the Epilepsies (Internet). 4th ed.
An important test to confirm the monogenetic nature of disease is to question whether the introduction of the same mutation into a mouse causes the animal to suffer disease. This has been accomplished for mutations in the SCN1A, HCN1, and GABARG2 genes, which each presents as epilepsy in mice. Prospective gene targeting in mice by genetic manipulation is laborious and not always successful. Considering the age dependence of epilepsy and species differences between humans and mice is important. The inability to reproduce a seizure phenotype by introducing
a disease-causing mutation in a mouse does not prove lack of causality. It is also important to realize that collectively these mutations account for only a very small percentage of epilepsy cases and that for many affected patients gene changes that would predict disease remain elusive. With regard to complex mutigenetic epilepsies, we are just beginning to see positive results from large studies conducted by consortia of investigators such as the European EPICURE Consortium.13 Such studies look for common genetic variations or
I. STATIC NERVOUS SYSTEM DISEASES
66
3. Seizure Disorders and Epilepsy
single-nucleotide polymorphisms. (For a better understanding of these genetic screens, the reader is directed to Chapter 16 where genome-wide association studies are explained in further detail). To be successful, such studies must include data from large patient cohorts. Thus far, increased susceptibility has been shown to correlate with variations in the SCN1A and CHRM3 genes, a Na+ channel and muscarinic receptor, respectively, and for three genes (VRK2, ZEB2, and PNPO) for which the link to epilepsy is less clear at the moment. With regard to multigenetic e pilepsies, it is important to stress that a combination of small changes in affected genes gives rise to a phenotype that may confer heightened susceptibility that only manifests as seizures under the appropriate environmental conditions, such as stress or hypoxia.
4.8 Emerging Targets and Mechanisms Suspected to Play a Role in Epilepsy KCC2/NKCC1 The importance of GABA as an inhibitory transmitter in the E–I balance has already been discussed. Being primarily a ligand-gated Cl− channel, the inhibitory action of GABA is due to the electrochemical gradient for Cl− across the plasma membrane. Cl− accumulates via the sodium–potassium–chloride transporter NKCC1 that imports Cl− together with Na+ into the cell. In mature neurons, NKCC1-mediated Cl− uptake is counteracted by a second transporter, the potassium chloride cotransporter KCC2, which effluxes Cl− (Figure 6). From this balance of import and
export, most mature neurons have a very low intracellular Cl− concentration of only 7–10 mM, whereas extracellular (Cl−) is much higher, around 140 mM. As this gradient favors the influx of Cl− upon opening of the GABA-gated channel, a negative charge enters the cell, thereby hyperpolarizing the membrane. During brain development, however, KCC2 expression is delayed and is not fully expressed until well after birth. Therefore, in the newborn (human or mouse), intracellular Cl− is considerably higher and GABA receptor activation leads to Cl− efflux rather than influx, thereby actually depolarizing the membrane and contributing to excitability.15 This developmental shift is graphically illustrated in Figure 6. This shift in E–I balance is at least one reason why newborns are generally more susceptible to seizures, a condition that rectifies itself as KCC2 expression increases in the ensuing weeks of postnatal life, during which seizures spontaneously disappear. While there currently is no feasible way to induce a premature expression of the KCC2 transporter, there is a potent drug to inhibit the NKCC1-mediated Cl− uptake. The loop diuretic bumetanide (Bumex), used to treat edema associated with heart failure or liver disease, has been tested in a clinical trial for severe infantile seizures with the rationale that it alters the intracellular Cl− concentrations (www.clinicaltrials.gov, identifier NCT00830531). A change in KCC2 expression and function also characterizes adult forms of human epilepsy,12 and a recent clinical study successfully used bumetanide to treat patients with MTLE.16 It is worth mentioning that bromide, which was used quite effectively to treat epilepsy in the past, may have
FIGURE 6 Cl− transport affects the action of γ-aminobutyric acid (GABA) as either an inhibitory or excitatory transmitter. In the developing, immature brain (B), intracellular Cl− is accumulated to a high concentration via the activity of the highly expressed NKCC1 transporter. As a result, GABA acts as an excitatory neurotransmitter (A). Upon maturation, NKCC1 expression is reduced and KCC2 expression now predominates shunting Cl− out of the cell. The resulting low intracellular Cl− makes GABA an inhibitory neurotransmitter. Reproduced with permission from Ref. 14.
I. STATIC NERVOUS SYSTEM DISEASES
4 Disease Mechanism/Cause/Basic Science
done so by affecting the transmembrane Cl− gradient as well. Bromide, like Cl−, is a monovalent halide anion, and it behaves much like Cl− in biological systems. It can therefore substitute as a charge carrier for the GABA receptor and is about 50% more permeable than Cl−. At an extracellular concentration of 10–20 mM, bromide enhances the flux of a negative, inhibitory charge into the cell compared with Cl− alone by about 28–35%. So without any other changes, the same amount of GABA increases inhibition by about one-third.17 Mammalian Target of Rapamycin Mammalian target of rapamycin (mTOR) is a protein kinase involved in regulating cell growth, proliferation, and survival and has been extensively studied in the cancer literature. The mTOR inhibitor rapamycin is used in clinical trials for the treatment of endometrial cancer and soft-tissue and bone sarcomas. Tuberous sclerosis is an inherited disorder with benign tumors and cortical thickenings called tubers. These are caused by the aberrant migration of neurons and cellular growth that presumably occur as a result of defective mTOR signaling. Recordings in patients determined that some of these tubers are intrinsically epileptogenic. Interestingly, by harboring the mutations that cause tuberous sclerosis, the mTOR inhibitor rapamycin suppressed seizures. It was also recently discovered that mTOR activity is regulated by pathways that are involved in glutamatergic, excitatory transmission and seems to be strongly associated with other forms of epilepsy. Because of the pleiotropic effects of mTOR signaling, its mechanistic involvement in epileptogenesis, that is, the development of epilepsy, is not yet known. However, in a number of preclinical epilepsy models, such as kainate-induced status epilepticus, rapamycin was able to suppress seizure development. Synaptic Vesicle Protein 2A Synaptic vesicle protein 2A (SV2A) is an abundant membrane-bound glycoprotein that is found on secretory vesicles including synaptic vesicles. It is an essential protein; mice with homozygote deletion die within 3 weeks after birth. SV2A has been identified as the binding site for the AED levetiracetam, although the mechanism whereby the drug reduces seizures is not understood. Interestingly, patients suffering from MTLE show a 32% decrease in SV2A expression,18 and a similar decrease in SV2A was observed in rodents following experimental induction of seizures. Moreover, the genetic deletion of the SV2A gene in neurons results in a decrease in GABA release in the hippocampus. This renders these animals more susceptible to seizures, although it is insufficient to induce epilepsy. In spite of its unclear role in epilepsy,
67
SV2A is an attractive target for the development of future AEDs. Neurotrophins Neurotrophins are important modulators of cell survival and development; some scientists consider them the “multivitamin of the brain.” The biology of BDNF is extensively discussed in Chapter 16. Four neurotrophins are expressed in the mammalian brain, including nerve growth factor, BDNF, and the neurotrophins 3 and 4. They each bind to distinct receptors called TrK receptors, tyrosine kinases that exhibit distinct affinity for select neurotrophins. TrKA binds nerve growth factor, TrKB is specific for BDNF and neurotrophin 4, whereas TrKC is activated by neurotrophin 3. Upon ligand binding, TrK receptors dimerize to become catalytically active. Numerous research studies support an important role for TrKB receptors activated via BDNF in epileptogenesis, although no single common mechanism has been elucidated.19 Elimination of TrkB in mice eliminates epileptogenesis in the kindling model of epilepsy, and overexpression of BDNF causes spontaneous seizures in mice and in brain slices. BDNF also encourages the sprouting of dendrites in the hippocampus, which contributes to aberrant connectivity. Finally, TrKB activation causes a reduction in KCC2 expression, which, as discussed above, leads to reduced GABAergic inhibition. Ca2 +-Regulated Enzymes Neuronal activity almost always causes increases in intracellular Ca2 +. This in turn can activate a number of Ca2 +-activated enzymes. Two of these, CaMKII and calcineurin, have received prominent attention in the epilepsy literature. Ca2 + causes a conformational change in CaMKII, activating its catalytic activity and causing phosphorylation. Mice with a mutated and nonfunctional CaMKII show limbic epilepsy, arising from the limbic structure of the brain for example the frontal lobe, and kindled mice show a reduction in CaMKII activity. Currently available data18 suggest that reduced CaMKII activity alone is sufficient to cause cellular and molecular changes that make the brains of mice epileptic. Calcineurin is a phosphatase; when activated by Ca2 + binding, calcineurin dephosphorylates its substrates. Calcineurin can be pharmacologically inhibited by the immunosuppressant drugs cyclosporine A or FK506, and treatment of mice with these drugs makes it much more difficult to induce seizures in mice through kindling. Increased activity of calcineurin, on the other hand, accompanies febrile seizures in infants. Both CaMKII and calcineurin are also important in learning and memory; their activity regulates Glu and GABA receptor trafficking and thereby alters the number of receptors in the membrane.
I. STATIC NERVOUS SYSTEM DISEASES
68
3. Seizure Disorders and Epilepsy
Astrocytes While it is undisputed that seizures are generated by the synchronous discharge of neuronal networks, nonneuronal cells may be complicit elements that support the genesis of seizures. Astrocytes have received most interest in this regard; these cells serve a multitude of functions to support normal neuronal activity. Astrocytes clear K+ and Glu released from neurons, and the loss of either Kir4.1, the channel responsible for K+ clearance, or EAAT2, the transporter that removes Glu, causes seizures in mice. Astrocytes catabolize Glu to Gln, which can serve as a precursor for the neuronal synthesis of GABA via glutamic acid decarboxylase. Astrocytes are also highly interconnected through gap junctions, allowing them to quickly shuttle molecules removed from the extracellular space to neighboring cells. Astrocytes show structural and functional changes when injured or when associated with disease, which makes them reactive. Such reactive astrocytes may no longer be able to perform their expected housekeeping roles and may indirectly contribute to disease. For example, reactive astrocytes may no longer import Glu from the extracellular space and convert it to Gln. The consequences may be dramatic, as the buildup of Glu near synapses chronically activates neuronal Glu receptors. Moreover, since astrocytes no longer supply neurons with a needed substrate for GABA synthesis, they starve neurons of GABA. Both increased extracellular Glu and decreased neuronal GABA upset the normal E–I balance. Reactive astrocytes also have been shown to produce and release proinflammatory cytokines including interleukin (IL)-1b and tumor necrosis factor-α. These molecules increase neuronal excitability and activate the adaptive immune system, both of which are undesirable effects in the setting of epilepsy. Stem Cells The formation of aberrant projections that is observed, for example, in the hippocampus following seizures suggests that structural plasticity contributes to a rewiring of neural networks. The discovery that the brain harbors neural stem cells throughout life prompted inquiry as to their participation in nervous system diseases and injury. Interestingly, the hippocampus, a highly seizure-prone brain region, also contains one of the niches where stem cells reside. The dentate gyrus produces neural stem cells that migrate into the hippocampus in a process called neurogenesis, and many of these adult-born neurons differentiate into GABAergic interneurons. Their contribution to abnormal networks has been documented in experimental models of epilepsy where the induction of status epilepticus was followed by marked neuronal cell death in the hippocampus and then by robust proliferation of progenitor
cells. In the ensuing 1–3 weeks, these newborn neurons contributed to aberrant networks in the hippocampus that showed abnormal processes and were placed in the wrong parts of the hippocampus, suggesting that they contribute to epileptogenesis.20 Epigenetics One of the most puzzling features of epilepsy is the unpredictable sporadic manifestation and the long delays that can exist between seizures. Moreover, a suspected brain insult such as trauma or infection is often followed by a long latent period until the first seizure presents. We generally believe that this epileptogenesis period wires the brain to have a heightened intrinsic susceptibility to trigger epileptic events. The emerging field of epigenetics describes a plausible way in which environmental factors control gene expression. Chapter 16 contains a much more elaborate discussion of epigenetics as it pertains to neurological disorders. In a nutshell, environmental factors can alter the accessibility of groups of genes on the chromosomes or of individual genes for transcription into proteins. A few proteins that are under environmental control thereby become the master keys to parts of the genome. These proteins are enzymes that attach or detach methyl or acetyl groups to cytosine or adenine nucleotides on DNA. Methylated DNA is typically silenced and not transcribed. Methylation occurs at so-called CpG dinucleotides or CpG islands. Following status epilepticus, over 300 genes are hypomethylated (reduced methylation), and consequently, gene expression is expected to be enhanced.21 By contrast, patients with MTLE show enhanced expression of the DNA methyltransferase gene, which would be expected to lead to enhanced transcriptional silencing of DNA. If the histones, which coil stretches of DNA, are methylated or demethylated, the transcription of many genes contained on a larger stretch of DNA is regulated. The master keys that do so include histone acetyltransferase, deacetylases, and histone methyltransferases. These are emerging pharmacological targets in epilepsy and other neurological disorders, particularly since two drugs (Vidaza and Dacogen) that target all DNA methyltransferases and inhibit DNA methylation are already clinically approved.22 Interestingly, one of the oldest AEDs, sodium valproate, is also an inhibitor of one of these (histone deacetylase), which when inhibited leads to uncoiling of DNA from histones and therefore generally enhances transcription. Hence, valproate may act not only as a regulator of neuronal ion channels but also as a master key regulating transcription of proteins that contribute to a number of chemical and structural changes, including neurogenesis, that occur during epileptogenesis.
I. STATIC NERVOUS SYSTEM DISEASES
5 Treatment/Standard of Care/Clinical Management
5 TREATMENT/STANDARD OF CARE/ CLINICAL MANAGEMENT 5.1 Diagnosis During a routine neurological examination, the treating neurologist gathers the patient’s history and any evidence of recent lifestyle changes, use of medication or drugs, injuries, diseases, or exposure to drugs or stress factors such as divorce or loss of a loved one. EEG evaluation is advisable. Given the infrequent occurrence of seizures in most patients, ambulatory, wearable EEG systems can be used on an outpatient basis. The EEG cannot always establish or refute the diagnosis of epilepsy, but if abnormalities are detected, it can aid in classifying the disease and may aid in determining the most appropriate drug regimen. Most patients will also be imaged by computed tomography or magnetic resonance imaging (MRI), if possible, to establish whether a structural abnormality, such as a tumor or cortical malformation, may contribute to the disease.
5.2 Pharmacological Treatment There are currently over 20 drugs in use as primary treatments for epilepsy (Table 2). These are typically referred to as antiepileptic drugs or AEDs. In general, they act to raise the threshold for neuronal network excitability. In light of what we learned above, there is a diverse group of proteins involved in neuronal activity, ranging from ion channels to neurotransmitter receptors and their transporters. Not surprisingly, different AEDs target different proteins within this network and the drugs often have pleiotropic effects, meaning that a single drug can affect multiple targets or pathways. The latter may be surprising, but as we learn in Chapter 17 of this book, few clinically used drugs were developed specifically to target known receptors but instead resulted from serendipitous discovery, with the drug existing long before its application was known. Table 2 summarizes the presumed mode of action and preferred use of current AEDs. As AEDs are supposed to enhance inhibition, it is not surprising that many drugs work by augmenting the GABA response (benzodiazepines and barbiturates) or the availability of GABA through reuptake inhibitors (gabapentin and tiagabine) or enhancing GABA release from synaptic vesicles (levitracetam). An overall reduction in Na+ channel activity that drives APs is another target of some widely prescribed drugs such as phenytoin (Dilantin), carbamazepine (TEGretol), lamotrigine, topiramate, and lacosamide. Finally, the release of neurotransmitters can be regulated by inhibition of certain Ca2 + channels (phenytoin, gabapentin, and pengabalin); ethusuximide and valproic acid presumably are effective in absence epilepsy by means of inhibiting T-type Ca2 + channels. Given their
69
different modes of action, some drugs can be used in combination with synergistic (additive) effects. The first line of treatment for most patients with focal epilepsy is carbamazepine, lamotrigene, phenytoin, or topiramate. Valproate is the primary choice and “Swiss army knife” for a number of generalized seizures, including absence seizures as well as myoclonic and atonic seizures. It is the most widely used drug worldwide to treat epilepsy. For one-third of patients the first drug choice provides freedom from seizures. However, if even after careful increase in drug dose a therapeutic response cannot be achieved, a second drug is added. Some patients must take three or four drugs simultaneously, and considering how they might interact is important. Note that epilepsy is considered medically refractory if two or more first-line AEDs have failed to provide lasting improvements. The goal of effective treatment is freedom from seizures. It is important to note that freedom from seizures does not necessarily equate to a completely normal EEG, and there is continued debate between scientists and clinicians regarding the value of considering EEG changes in the assessment of the efficacy of a certain treatment approach. From a purely clinical point of view, “we are not treating the EEG but the seizure” is a frequently voiced opinion. However, since recent research suggests that the masking of behavioral seizures in the continued presence of underlying EEG abnormalities may signify a continued deterioration of brain function, including cognitive decline, this opinion may warrant being revisited. With the currently available drugs, approximately twothirds of patients are effectively managed, with the majority of patients achieving freedom from seizures for most of their life. For the remaining one-third of drug-resistant or medically refractory patients, alternative strategies must be considered. Foremost, surgical approaches are particularly effective in a subgroup of epilepsy syndromes.
5.3 Surgical Treatment Patients who do not respond to any of the AEDs often have a poor quality of life and an inability to work or even live independently. For some of these patients, particularly those with “lesional epilepsy” who are harboring structural brain abnormalities that give rise to seizures, surgery can be an effective alternative. Before surgery, a meticulous neurological evaluation including EEG and MRI must be performed. Structural changes are often subtle and require high-field-strength (3T) images using a combination of T1 and T2 and possibly fluid-attenuated inversion recovery sequences to visualize abnormalities. It is essential that multiple tests converge on the same seizure focus to ensure that the offending part of the brain can be surgically removed while sparing as much of the surrounding brain as possible. During surgery, focal cortical stimulation while the patient is awake and naming objects displayed
I. STATIC NERVOUS SYSTEM DISEASES
70
3. Seizure Disorders and Epilepsy
TABLE 2 Some currently available antiepileptic drugs (in alphabetical order). Drug name
Trade names
Presumed mechanism of action
Target
−
Indication
Comments No longer used in humans, still available for veterinary use
Bromide salt
Bromo-Selzer
Unknown
May replace Cl as permeable ion at GABA-R; has higher permeability than Cl−
Tonic–clonic seizures
Carbamazepine
Tegretol
Voltage-gated Na+ channels
Reduces repetitive neuronal discharge, anticonvulsant
Focal and generalized seizures of all types
Ethosuximide
Zarontin
T-type Ca2 + channels
Targets Ca2 + channels in thalamic neurons involved in generation of aberrant spike– wave discharges
Absence epilepsy
Ezogabine
Ezogabine
K+ channels, KCNQ Kv7.2
Opens K+ channels, stabilizes membrane potential, reduces excitability
Focal seizures
Felbamate
Felbatol
NR2B subunit of NMDA-R, Na+ channels, GABA-R
Open channel blocker, inhibits Focal seizures, NMDA, thereby reducing Lennox–Gastaut glutamatergic excitatory syndrome activity
Gabapentin/ pregabalin
Neurontin
Ca2 + channels, GABA Unclear, increases synaptic Focal seizure transporters? GABA concentrations, thereby enhancing inhibition
Lacosamide
Vimpat
Na+ channels
Slows recovery from inactivation
Lamotrigine
Lamictal
Ca2 + channels, Na+ channels
Blocks voltage-gated Na+ Tonic–clonic focal and channels, reduces the release of generalized seizures; glutamate Lennox–Gastaut syndrome
Levetiracetam
Keppra
Synaptic vesicle protein 2A
Changes release of neurotransmitters
Focal seizures, myoclonic, tonic–clonic
Oxcarbazepine
Trileptal
Na+ channels, N-type Ca2 + channels
Stabilizes resting membrane potential
Focal seizures
Analog of carbamazepine
Perampanel
Fycompa
AMPA-R
Noncompetitive agonist of AMPA Glu receptors, reduces excitability
Refractory, focal seizures
High potency, long systemic bioavailability
Phenobarbital
Luminal
GABA-A receptor
Enhances GABAergic inhibition
Phenytoin
Dilantin
Voltage-gated Na+ channels
Use-dependent inhibitor, becomes more effective with repetitive neuronal activity
Rufinamide
Banzel
Na+ channels
Maintains Na+ channel in closed, inactive state
Tiagabine
Gabitril
GABA transporter
Inhibit reuptake of GABA, Focal seizures, adjunct therefore prolonging inhibitory treatment action of GABA
Topiramate
Topamax
AMPA-R, GABA-R, Na+ channel
Multiple effects proposed, Lennox–Gastaut inhibitor of AMPA Glu syndrome, focal receptors and blocker of Na+ seizures channels reduce excitatory drive, also enhanced GABA-Rmediated inhibition
No known interactions with other AEDs Enzyme inducer that interacts with several other AEDs, rare cause of liver failure Originally designed to mimic GABA but does not work through GABA-R, pregabalin is the successor and is more potent, has better bioavailability
Focal seizures
Enzyme inducer
Adjunct for Lennox– Gastaut syndrome
I. STATIC NERVOUS SYSTEM DISEASES
Inhibits carbonic anhydrase, increases risk for kidney stones, enzyme inducer, interacts with other AEDs
71
5 Treatment/Standard of Care/Clinical Management
TABLE 2 Some currently available antiepileptic drugs (in alphabetical order)—cont’d Presumed mechanism of action
Drug name
Trade names
Target
Indication
Comments
Valproate
Depakote
Multiple, GABA-R, GABA transaminase, NMDA-R, HDACs
Reduces glutamatergic excitation via NMDA-R, enhances GABA inhibition, changes transcription of multiple genes via HDAC inhibition
Absence seizures, Introduced in 1967, now tonic–clonic, the most widely prescribed focal seizures AED worldwide with dyscognia, posttraumatic epilepsy, Lennox–Gastaut syndrome
Vigabatrin
Sabril
GABA transaminase
Irreversible inhibitor of the enzyme that catabolizes GABA; drug increases available GABA and enhances inhibition
Focal and secondary generalized seizures, infantile spasms, Lennox–Gastaut syndrome
Zonisamide
Zonegran
Na+ and Ca2 + channels Unknown mechanism of action, primary target is carbonic anhydrase but likely reduces excitability by interacting with ion channels
Negative interactions with carbamazepine and phenytoin reported
Focal and generalized tonic–clonic seizures
AED, antiepileptic drug; AMPA-R, AMPA receptor; GABA, γ-aminobutyric acid; GABA-R, γ-aminobutyric acid receptor; Glu, glutamate; HDAC, histone deacetylase; NMDA-R, N-methyl-d-aspartate receptor. Compiled using data from: French G. Continuum, Lifelong Learning in Neurology – Epilepsy. June 2013;19(3).
on a computer screen allow the surgeon to further map the “eloquent cortex,” those areas of the cortex essential for language and motor and sensory function. The most common surgically curative lesional epilepsy is mesial-temporal lobe epilepsy (MTLE), for which complete or partial removal of the anterior temporal lobe, along with partial resection of the hippocampus and amygdala on the affected hemisphere, provides excellent outcome in many patients, with approximately 60% being seizure free after 1 year. An example of a patient before and after surgery is illustrated in Figure 7. Dissection of the corpus callosum provides benefit to some of the severe, medically intractable forms of e pilepsy, for example, Lennox– Gastaut syndrome, a severe form of childhood epilepsy often associated with visible cortical malformations. A number of acquired epilepsy syndromes benefit from surgical intervention. Most notable are benign or low-grade brain tumors, including gangliogliomas and oligodendrogliomas, which can give rise to chronic epilepsy. Similarly, traumatic brain injury can leave patients with chronic seizures, and removal of scar tissue surrounding the lesion has the potential to ameliorate the disease. Another surgical indication is seizures resulting from vascular malformations that can in some instances be surgically corrected. The most drastic surgery that is routinely performed involves the removal of large sections of a brain hemisphere in children suffering from Rasmussen encephalitis, a rare, debilitating, and medically intractable epilepsy presumably caused by an autoimmune response to brain Glu receptors. As drastic as this intervention sounds, it carries a high chance of controlling seizures while maintaining overall acceptable neurological function.
Overall, surgical successes have improved considerably in recent years, particularly with the introduction of preclinical mapping of eloquent brain regions and careful intraoperative stimulation mapping of the seizure focus. Almost 70% of patients with MTLE and 50% of patients with focal neocortical resections report freedom from seizures after surgery.24 However, because of the understandable reluctance of patients to undergo surgery, it is pursued in far fewer patients than are likely to benefit and is often only considered after a patient has already suffered uncontrolled seizures for many years.
5.4 Neuromodulatory Epilepsy Therapy In patients who are medically refractory and surgery is not an option, neuromodulatory approaches should be considered.25 These are particularly useful in patients where seizures originate from multiple sites or from eloquent areas of the brain that cannot all be resected. Although different approaches are discussed below, they all seek to dampen neuronal excitability. Some are openloop devices that provide simple electrical stimulation such as deep brain stimulation (DBS), trigeminal nerve stimulation (TNS) and vagal nerve stimulation (VNS), while others such as responsive neural stimulation (RNS) use closed-loop systems that detect seizure activity and only inject current when epileptiform activity is detected. Neuromodulatory devices are advantageous compared to medication and surgery, in that they have minimal, if any, side effects and are far less invasive compared to surgery. Also, in many instances, their efficacy increases over time. In the United States only VNS and
I. STATIC NERVOUS SYSTEM DISEASES
72
3. Seizure Disorders and Epilepsy
(A)
(B)
(C)
(D)
FIGURE 7 Surgical treatment of epilepsy. Coronal magnetic resonance image (MRI; fluid-attenuated inversion recovery) of brain before surgery in coronal (A) and axial (B) view before surgery showing right mesial-temporal sclerosis (arrows) with compensatory dilatation of the temporal horn before surgery. (C) and (D) show the same views, respectively, of T1-weighted MRI scans images after surgical removal of the amygdala, hippocampus, and anterior temporal lobe. From Ref. 23.
RNS devices are FDA approved specifically for the treatment of epilepsy, yet others are approved in Europe and other parts of the world and can be used “off-label” at the physician’s discretion. Vagal nerve stimulation (VNS): Approved by the FDA in 1997, VNS is the oldest open-loop neurostimulation approach. The electrode is surgically implanted around the left vagus nerve in the subclavicular region. The stimulator generates mild, regular pulses that travel via the vagus nerve and provide pacemaker-like activity that attenuates seizure activity. The stimulator can be turned on and off by the patient through a magnet swiped over the skin. VNS is most efficacious for generalized seizures and in clinical trials provides > 50% seizure reduction. Trigeminal nerve stimulation (TNS): TNS is approved in Europe to treat partial seizures but is often used “off- label” for treatment in the United States. TNS is a noninvasive therapy whereby the electrodes are placed on the skin above the supraorbital branches of the trigeminal nerve. Its mechanism is similar to VNS, and TNS results in up to a 35% reduction in seizure frequency with minimal to no side effects. Deep brain stimulation: While the FDA has not approved DBS for the specific treatment of epilepsy, it is approved in Europe and Canada and used widely
off-label in the United States. The depth electrodes and implanted stimulator are similar to the device extensively discussed in Chapter 5. However, for treatment of epilepsy, the electrodes target the anterior portion of the thalamus, a major relay station in the underlying cortical–hippocampal–thalamic seizure circuit. Excellent results have been obtained for temporal lobe epilepsy with up to 70% seizure reduction after 5 years of use. Responsive neurostimulation (RNS): RNS is a closedloop stimulator that was approved by the FDA in 2013 for patients with partial seizures originating from multiple identified foci. Strips of multipolar recording electrodes are implanted under the dura, the outer most protective layer covering the brain, and monitor the cortex for epileptiform activity. When abnormal activity is detected, an intracranially implanted stimulator triggers trains of preprogrammed electrical pulses that seeks to dampen seizure activity, akin to a “noise-canceling” headset eliminating unwanted sound. Clinical studies show long-term use (6 year) providing 65% reduction in seizure frequency with only minimal side effects. Noninvasive real-time seizure alerting smart watch: While not a treatment, it is worthy to mention another recent electronic device approved by the FDA in 2018, that shows promise in the management of epilepsy in children
I. STATIC NERVOUS SYSTEM DISEASES
5 Treatment/Standard of Care/Clinical Management
6 years or older. The Embrace smartband (Empatica Inc.) is the first device that monitors the body for seizure activity without the use of EEG through interpretation of a number of physiological signs including variabilities in heart rate, skin conductivity, skin temperature, acceleration, and movement. The wristband provides warning of an impending or acute generalized tonic–clonic seizures with a high sensitivity (93%) and very low false positives (0.6%). It alerts caregivers of a child having an event, for example, during sleep that could result in sudden unexplained death with epilepsy (SUDEP).
5.5 Immunological Approaches, Diet, and Alternative Treatments It is abundantly clear that seizures can trigger various immune responses in the brain. Although typically assumed to be immune-privileged, the brain does have some semblance of both activated and innate immunity. Specifically, microglia, astrocytes, and endothelial cells can become activated and release a number of inflammatory molecules including IL-1β, TNF, IL-6, prostaglandin E2, and complement. Some of these cytokines are released as a consequence of fevers. IL-1β, for example, is released into the brain by elevating temperature alone, and it is possible that IL-1β plays a role in febrile, temperature-induced seizures in infants. However, currently, there is not sufficient evidence to suggest causality for cytokines in the generation of seizures, let alone epilepsy. Autoantibodies to specific nervous system proteins, for example, GluR3, are present in Rasmussen encephalitis, a catastrophic form of childhood epilepsy that presents with brain atrophy. These brains also contain reactive astrocytes and activated microglial cells. Antibody removal by plasmapheresis or intravenous immunoglobulin as well as reducing inflammation with corticosteroids or corticosteroid-releasing hormones has been explored for the treatment of some of these epilepsies, yet with inconclusive effectiveness. This leaves us to wonder whether the immune response was a consequence of the seizure or whether it may have been the cause. At present these treatments are not generally recommended.26 Ketogenic diet: By contrast, an effective alternative treatment pursued by some patients is the ketogenic diet, a diet rich in fat and low in carbohydrates. It is probably the most well documented and longest established. It is based on the serendipitous finding some decades ago that prolonged periods of fasting reduce the frequency of seizures in some epileptic patients. The body typically uses glucose as its main source of energy. Glucose is stored in the form of glycogen in the liver and other organs. The total stored glucose sustains the body for about 24 h. After that, the body begins using fat as an energy substrate, and the liver converts fat to fatty acids
73
or ketone bodies that give this diet its name. Ketone bodies are used as an energy substrate in place of glucose in the brain. The classic ketogenic diet consists of 80% fat, the balance being glucose and protein. The ketogenic diet is particularly interesting in the treatment of refractory epilepsy in children and adolescents. A recent randomized controlled clinical study provides strong evidence that the addition of ketogenic diet to standard AEDs reduces seizure frequency and severity by about 50% after 4 months.27 Even more striking, children with Dravet syndrome, a severe form of childhood epilepsy typically caused by mutations in the Nav1.1 Na+ channel, showed 75% of children who remained on the ketogenic diet showing lasting reduction in the number of seizures experienced.28 However, long-term compliance with this diet is difficult, particularly in children. Ongoing studies are experimenting with a modified Atkins diet, which is easier to tolerate by permitting more carbohydrates and proteins. Note that we currently do not fully understand how the ketogenic diet reduces seizures. However, ketone bodies directly reduce neural excitability by activating ATP-sensitive K+ channels.29
5.6 Comorbidities and Psychological Issues to Consider Given the fact that seizures represent an unusual discharge of electrical activity in relatively large areas of the brain, frequently including brain regions associated with higher cognitive functions (cortex) or memory (hippocampus), it is no surprise that uncontrolled seizures present with a significant decline in cognitive and memory function. Indeed, approximately 75% of newly diagnosed and untreated adults with epilepsy have deficits in attention, executive function, and memory. Significantly, cognitive impairment in patients with newly diagnosed epilepsy surpasses that seen in the early stages of other cerebral diseases such as Parkinson Disease or multiple sclerosis.30 Memory decline can be associated with a loss of hippocampal volume that is visible on MRI. An important consideration with regard to patients treated with AEDs is the potential direct effect that these drugs may have on cognitive and memory function as well as mood and behavior. Since AEDs interfere with both excitatory and inhibitory neurotransmitter systems, which are fundamentally involved in information processing, it is not a surprise that chronic AED treatments can alter cognitive function. Of the available drugs, barbiturates, benzodiazepines, and topiramate seem to cause the most negative cognitive side effects. Up to 50% of patients with medically refractory epilepsy and about 10% of patients with well-controlled seizures present with anxiety and/or depression. This is not surprising given that patients report a feeling of helplessness and loss of control as they become
I. STATIC NERVOUS SYSTEM DISEASES
74
3. Seizure Disorders and Epilepsy
i ncreasingly dependent on others and are often unable to live independently, hold a steady job, or even drive. Yet depression is the most underdiagnosed and untreated condition in patients with epilepsy and must be adequately addressed in a comprehensive treatment plan by, for example, offering selective serotonin reuptake inhibitors in conjunction with AEDs. Another important consideration regarding the medical control of seizures is how AEDs may affect fetal development if an expecting mother with seizures uses AEDs. Since significant brain development occurs before a prospective mother knows of her pregnancy, care must be taken to avoid potentially harmful exposure of the fetus. This is particularly relevant for valproate, which, although highly effective in a number of seizure disorders, also is a teratogen, that is, a drug that can lead to significant birth defects. Valproate has been reported to significantly lower a child’s mean intelligence quotient from 100 to 92 by the age of 3 years, with follow-up showing impaired verbal, nonverbal, and memory function by the age of 6 years. As a consequence, the American Academy of Neurology recommends avoiding valproate in women during pregnancy or women likely to become pregnant.
6 EXPERIMENTAL APPROACHES/ CLINICAL TRIALS Until we can provide effective relief and achieve seizure control for all patients, we must continue to explore new treatment strategies. Following laboratory discoveries and preclinical testing in animal models of disease, new approaches are subject to controlled experimentation in human clinical trials. Active clinical trials are published for public access on www.clinicaltrials.gov, a searchable database that informs patients about their potential eligibility to participate in a clinical trial. The process of clinical experimentation is expanded upon in Chapter 17 of this book. Typically, early-stage trials (phase I/II) are exploratory and examine safety, whereas later-stage clinical trials (phase III/IV) are more mature and seek to show enhanced efficacy over other available treatment options. As of September 2020, there are about 380 clinical epilepsy trials, of which 121 are designated interventional studies. Of these, 76 are early (phase 0–II) studies and 58 are phase III or IV. The remaining studies are observational or examine exercise, diet, or even stress management. Others look at comorbidities such as vision effects or decline in cognitive function. Most pharmacological studies are examining the efficacy of some second-generation AEDs that have already passed previous clinical trials, including lacosamide, brivaracetam (an analog of levetiracetam), lamotrigine, and retigabine for different seizure indications or in different
c ombination therapies. For example, ezogabine is a K+ channel activator that had significant reduced seizure frequency in a previous phase II study of 399 participants with refractory partial seizures31 and is now being studied in combination with other established AEDs. Lacosamide was first introduced for clinical use in 2009. It acts on voltage-gated Na+ channels and enhances slow inactivation of the channel. In a placebo-controlled study, lacosamide significantly reduced the number of partial seizures. It is now being studied in intravenous form as opposed to oral dosing, and another study examines the drug’s effect on sleep–wake cycles. With regard to new antiepileptic drugs currently in clinical evaluation,32 a few are worth mentioning: Cannabidiol (CBD) is the second most abundant cannabinoid found in the Cannabis sativa plant. Unlike tetrahydrocannabinol (THC), CBD does not bind significantly to cannabinoid receptors and hence lacks the psychoactive effect of THC. Following a number of anecdotal reports, a highly purified medical formulation of CBD (Epidiolex), was clinically evaluated over a 14-week period in children suffering from Dravet syndrome (www. clinicaltrials.gov, identifier NCT02091375). The trial used a daily dose of 500–1500 mg, doses much more potent than CBD sold over the counter (typical daily over the counter doses are 10–50 mg). The results were promising, with frequency of convulsive seizures decreasing significantly from an average of 12/day to 6/day. There was no reduction in nonconvulsive seizures. Epidiolex was subsequently studied in a phase II study in children with Lennox–Gastaut syndrome (www.clinicaltrials.gov, identifier NCT02224690), a severe form of childhood epilepsy with multiple and often concurrent seizures daily, that develop in children aged 3–5 and persist into adulthood. Epidiolex provided equally promising results with a 44% reduction in seizures on CBD compared to 22% in the placebo group. Comparing the relative effect size of CBD versus placebo in both of the above studies suggests that CBD improves the outcome by about 20%. Note that these patients still suffer countless daily seizures and remain severely incapacitated. Nevertheless, based on these two studies, the FDA approved Epidiolex in 2018 for seizure treatment in patients with Dravet and Lennox Gastaut. Whether there is a broader benefit for other forms of epilepsy is currently unknown and the clinical literature generally discourages from using CBD to treat epilepsy. It must be stressed that the mechanism of action underlying the anticonvulsive effect of CBD is poorly understood. Given the fact that high doses of CBD slow the hepatic clearance of AEDs via inhibition of cytochrome P450, it is possible that CBD simply enhances concentrations and slows clearance of standard AEDs. Fenfluramine is a synthetic drug that interacts with several serotonin receptors and is being explored as adjuvant treatment in patients with Dravet syndrome.
I. STATIC NERVOUS SYSTEM DISEASES
7 Challenges and Opportunities
A phase III clinical trial recently reported a promising, ~ 64% greater reduction in motor seizures than placebo. Ganaxolone, a synthetic analog of the hormone allopregnanolone is an allosteric modulator of the GABAA receptor. The drug showed promise in a number of uncontrolled open-label studies across several seizure disorders. However, results of a recently concluded phase III trial in patients with focal seizures (www.clinicaltrials.gov, identifier NCT02358538) were disappointing with drug effects being equal to placebo. Padsevonil is the first example of a rationally designed antiepileptic drug that has multiple sites of action. Like one of the most common AEDs levetiracetam (Keppra), padsevonil binds to the SV2 presynaptic protein to reduce neurotransmitter release. Unlike Keppra, it binds to all three forms of SV2 (SV2A–SV2C). Furthermore, padsevonil also binds to the benzodiazepine site of the postsynaptic GABAA receptor enhancing GABAergic inhibition. Thus far, the drug has been evaluated only in patients suffering from unusually severe focal epilepsy where it has shown modest but clinically useful effects. A larger multicenter efficacy study (ARISE, www. clinicaltrials.gov, identifier NCT03370120) just reported (3/13/2020) disappointing results and a failure to achieve significant improvements compared to placebo. In addition to trials exploring the efficacy of drugs for established epilepsy, there are new trials examining whether prophylactic drug treatment may prevent epilepsy following a known insult, notably trauma. One such study uses biperiden, an atropine-like drug that works on muscarinic acetylcholine receptors and has previously been tested in patients with Parkinson Disease. Similarly, levetiracetam (Keppra) is being examined as an epilepsy prophylactic in patients who suffered hemorrhagic stokes.
7 CHALLENGES AND OPPORTUNITIES One hundred fifty years into the pharmacological management of epilepsy and the statistics are startlingly sobering. Even today, roughly one-third of all patients cannot be effectively treated and are forced to live with a significant disability and a reduced quality of life. Does that mean that we have not made progress? Of course not! The clinical management of epilepsy is vastly superior today than it was even 20 years ago. Extensive research has uncovered many pathways and gene candidates through which epilepsy can manifest. Outstanding animal models of disease now offer the opportunity to test new compounds in clinically relevant model systems. However, as is the case with other nervous system disorders, translating findings in animal models into effective treatment in humans is often a long and tedious process. Without question, we have learned
75
that epilepsies are a large collection of conditions that jointly manifest with similar electrophysiological and behavioral abnormalities, but they certainly do not define a single disease. The underlying root causes may be as varied as the conditions that cause a patient to run a fever. The principal tenet, however, that epilepsy results from an impaired E–I balance, is supported by an abundance of evidence. Given the many pathways through which this balance can be altered, there are tremendous opportunities to attempt to restore this balance through specific drugs. The greatest challenge remains the identification of new pathways that could yield unexploited drug targets to help those individuals who remain refractory to current pharmacological and surgical treatment. The emergence of second-generation drugs such as ezogabine, which reduce excitability through completely different mechanisms as the first-generation AEDs, are promising steps in this direction. Given the diverse AEDs available, it is possible—even likely—that personalized medicine holds the greatest promise in providing individual patients with a cocktail of drugs tailored to their specific epilepsy. This approach could be aided by further identification of genes associated with disease and the emergence of whole-genome sequencing, which will allow physicians to consider a person’s genetic makeup when selecting a drug cocktail best tailored to that patient. It could be argued that seizure disorders may offer an ideal testing ground to explore the implementation of personalized medicine. We have effective tools to diagnose the disease as well as accurately assess a patient’s drug response, without the need to wait for years. We also have numerous patient-specific presentations on which to try variants of different treatment protocols. Finally, and probably most important, there exist a significant market and unmet medical need for drug companies to serve. Drug discovery is ultimately driven by market forces, and companies will develop personalized drugs only if significant financial rewards exist; millions of patients worldwide who are suffering from drug-resistant epilepsy provide a viable market opportunity.
Acknowledgments This chapter was kindly reviewed by Christopher B. Ransom, MD, PhD, Director, Epilepsy Center of Excellence and Neurology Service, VA Puget Sound & Department of Neurology, University of Washington.
References 1. Hunt WA, Temkin O. The Falling Sickness: A History of Epilepsy From the Greeks to the Beginnings of Modern Neurology. Baltimore: Johns Hopkins Press; 1945. XV + 380. [J Clin Psychol. 1946;2:301. doi:10.1002/10974679(194607)2:3 3.0.CO;2-7]. 2. International League against Epilepsy. The history and stigma of epilepsy. Epilepsia. 2003;44:12–14. https://doi. org/10.1046/j.1528-1157.44.s.6.2.x.
I. STATIC NERVOUS SYSTEM DISEASES
76
3. Seizure Disorders and Epilepsy
3. Jackson J. A study of convulsions. Arch Neurol. 1970;22:184–188. https://doi.org/10.1001/archneur.1970.00480200090012. 4. Jackson JH. A study of convulsions. Trans St Andrews Med Graduates Assoc. 1870;3:162–204. 5. Friedlander WJ. The rise and fall of bromide therapy in epilepsy. Arch Neurol. 2000;57:1782–1785. 6. GBD 2016 Epilepsy Collaborators. Global, regional, and national burden of epilepsy, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurol. 2019;18:357. https:// doi.org/10.1016/s1474-4422(18)30454-x. 7. Sirven JI, Waterhouse E. Management of status epilepticus. Am Fam Physician. 2003;68:469–476. 8. Scharfman HE. The neurobiology of epilepsy. Curr Neurol Neurosci Rep. 2007;7:348–354. 9. Liu CH, Lin YW, Tang NY, Liu HJ, Hsieh CL. Neuroprotective effect of Uncaria rhynchophylla in Kainic acid-induced epileptic seizures by modulating hippocampal mossy fiber sprouting, neuron survival, astrocyte proliferation, and S100B. Evid Based Complement Alternat Med. 2012;2012:194790. https://doi. org/10.1155/2012/194790. 10. Goldberg EM, Coulter DA. Mechanisms of epileptogenesis: a convergence on neural circuit dysfunction. Nat Rev Neurosci. 2013;14:337–349. https://doi.org/10.1038/nrn3482. 11. Meisler MH, Kearney JA. Sodium channel mutations in epilepsy and other neurological disorders. J Clin Invest. 2005;115:2010–2017. https://doi.org/10.1172/jci25466. 12. Miles R, Blaesse P, Huberfeld G, Wittner L, Kaila K. In: Noebels JL, et al., eds. Jasper's Basic Mechanisms of the Epilepsies. National Center for Biotechnology Information (US); 2012. Michael A. Rogawski, Antonio V. Delgado-Escueta, Jeffrey L. Noebels, Massimo Avoli and Richard W Olsen. 13. Allen AS, et al. De novo mutations in epileptic encephalopathies. Nature. 2013;501:217–221. https://doi.org/10.1038/ nature12439. 14. Ben-Ari Y. Excitatory actions of gaba during development: the nature of the nurture. Nat Rev Neurosci. 2002;3:728–739. 15. Ben-Ari Y, Khalilov I, Kahle KT, Cherubini E. The GABA excitatory/inhibitory shift in brain maturation and neurological disorders. Neuroscientist. 2012;18:467–486. 16. Eftekhari S, et al. Bumetanide reduces seizure frequency in patients with temporal lobe epilepsy. Epilepsia. 2013;54:e9–12. https://doi. org/10.1111/j.1528-1167.2012.03654.x. 17. Suzuki S, et al. Bromide, in the therapeutic concentration, enhances GABA-activated currents in cultured neurons of rat cerebral cortex. Epilepsy Res. 1994;19:89–97. 18. de Groot M, Aronica E, Heimans JJ, Reijneveld JC. Synaptic vesicle protein 2A predicts response to levetiracetam in patients with glioma. Neurology. 2011;77:532–539. https://doi.org/10.1212/ WNL.0b013e318228c110. 19. McNamara JO, Huang YZ, Leonard AS. Molecular signaling mechanisms underlying epileptogenesis. Sci STKE. 2006;2006:re12. https://doi.org/10.1126/stke.3562006re12. 20. Parent JM, Lowenstein DH. Seizure-induced neurogenesis: are more new neurons good for an adult brain? Prog Brain Res. 2002;135:121– 131. https://doi.org/10.1016/s0079-6123(02)35012-x. 21. Miller-Delaney SF, et al. Differential DNA methylation patterns define status epilepticus and epileptic tolerance. J Neurosci. 2012;32:1577–1588. https://doi.org/10.1523/ jneurosci.5180-11.2012. 22. Kobow K, Blumcke I. The emerging role of DNA methylation in epileptogenesis. Epilepsia. 2012;53(Suppl. 9):11–20. https://doi. org/10.1111/epi.12031. 23. Chowdhury FH, et al. Microneurosurgical management of temporal lobe epilepsy by amygdalohippocampectomy (AH) plus standard anterior temporal lobectomy (ATL): a report of our initial five cases in Bangladesh. Asian J Neurosurg. 2010;5:10–18.
24. Spencer SS, et al. Predicting long-term seizure outcome after resective epilepsy surgery: the multicenter study. Neurology. 2005;65:912– 918. https://doi.org/10.1212/01.wnl.0000176055.45774.71. 25. Nune G, DeGiorgio C, Heck C. Neuromodulation in the treatment of epilepsy. Curr Treat Options Neurol. 2015;17:375. https://doi. org/10.1007/s11940-015-0375-0. 26. Vezzani A, French J, Bartfai T, Baram TZ. The role of inflammation in epilepsy. Nat Rev Neurol. 2011;7:31–40. https://doi.org/10.1038/ nrneurol.2010.178. 27. Lambrechts DA, et al. A randomized controlled trial of the ketogenic diet in refractory childhood epilepsy. Acta Neurol Scand. 2017;135:231–239. https://doi.org/10.1111/ane.12592. 28. Cross JH, et al. Dravet syndrome: treatment options and management of prolonged seizures. Epilepsia. 2019;60(Suppl. 3):S39–s48. https://doi.org/10.1111/epi.16334. 29. Giménez-Cassina A, et al. BAD-dependent regulation of fuel metabolism and K(ATP) channel activity confers resistance to epileptic seizures. Neuron. 2012;74:719–730. https://doi.org/10.1016/j. neuron.2012.03.032. 30. Witt JA, Helmstaedter C. Should cognition be screened in new-onset epilepsies? A study in 247 untreated patients. J Neurol. 2012;259:1727–1731. https://doi.org/10.1007/s00415-012-6526-2. 31. Porter RJ, Partiot A, Sachdeo R, Nohria V, Alves WM. Randomized, multicenter, dose-ranging trial of retigabine for partial-onset seizures. Neurology. 2007;68:1197–1204. https://doi.org/10.1212/01. wnl.0000259034.45049.00. 32. Bialer M, et al. Progress report on new antiepileptic drugs: a summary of the Fourteenth Eilat Conference on New Antiepileptic Drugs and Devices (EILAT XIV). I. Drugs in preclinical and early clinical development. Epilepsia. 2018;59:1811–1841. https://doi. org/10.1111/epi.14557.
General Readings Used as Source 1. Lowenstein DH. Seizures and epilepsy [chapter 369]. In: Longo DL, Fauci AS, Kasper DL, Hauser SL, Jameson J, Loscalzo J, eds. Harrison’s Principles of Internal Medicine. New York, NY: McGrawHill; 2012. 2. Pedley TA, Bazil CW, Morrell MJ. Meritt’s Textbook for Neurology. 10th ed. Philadelphia: Lippincott Williams & Wilkins; 2000. 3. In June of 2013 the journal Continuum: Lifelong Learning in Neurology published a special volume on Epilepsy by various authors. 19(3):551–879. 4. Pitkraenen A, Schwarzkroin A, Moshe A. Models of Seizures and Epilepsy. Elsevier [In: Noebels JL, Massimo A, Rogawski Michael A, Olsen Richard W, Delgado-Escueta Antonio V, eds. Jasper’s Basic Mechanisms of the Epilepsies (Internet). Bethesda, MD: National Center for Biotechnology Information (US)]; 2006. 5. McNamara JO, et al. Molecular signaling mechanisms underlying epileptogenesis. Sci STKE. 2006;2006(356):re12. 6. Thijs RD, Surges R, O’Brien TJ, Sander JW. Epilepsy in adults. Lancet. 2019;393:689–701. https://doi.org/10.1016/s0140-6736(18)32596-0. 7. A public guide to understanding genetic mutations can be found at: www.ghr.nlm.nih.gov (Valuable resource to search for genetic mutations by gene or protein name).
Suggested Papers or Journal Club Assignments Basic 1. Kearney JA, et al. A gain-of-function mutation in the sodium channel gene Scn2a results in seizures and behavioral abnormalities. Neuroscience. 2001;102(2):307–317.2.
I. STATIC NERVOUS SYSTEM DISEASES
REFERENCES
2. Viitanen T, et al. The K+–Cl cotransporter KCC2 promotes GABAergic excitation in the mature rat hippocampus. J Physiol. 2010;588(Pt 9):1527–1540. 3. Salmi M, et al. Tubacin prevents neuronal migration defects and epileptic activity caused by rat Srpx2 silencing in utero. Brain. 2013;136(Pt 8):2457–2473. 4. Giménez-Cassina A, et al. BAD-dependent regulation of fuel metabolism and K(ATP) channel activity confers resistance to epileptic seizures. Neuron. 2012;74:719–730. https://doi.org/10.1016/j. neuron.2012.03.032. 5. Hunt RF, Girskis KM, Rubenstein JL, Alvarez-Buylla A, Baraban SC. GABA progenitors grafted into the adult epileptic brain control seizures and abnormal behavior. Nat Neurosci. 2013;16(6):692–697. 6. Lemke JR, et al. Mutations in GRIN2A cause idiopathic focal epilepsy with rolandic spikes. Nat Genet. 2013;45:1067–1072. https:// doi.org/10.1038/ng.2728. 7. Sada N, Lee S, Katsu T, Otsuki T, Inoue T. Epilepsy treatment. Targeting LDH enzymes with a stiripentol analog to treat epilepsy. Science. 2015;347:1362–1367. https://doi.org/10.1126/science.aaa1299. 8. Lieb A, et al. Biochemical autoregulatory gene therapy for focal epilepsy. Nat Med. 2018;24:1324–1329. https://doi.org/10.1038/ s41591-018-0103-x. 9. Diaz Verdugo C, et al. Glia-neuron interactions underlie state transitions to generalized seizures. Nat Commun. 2019;10:3830. https://doi.org/10.1038/s41467-019-11,739-z. 10. Tewari BP, et al. Perineuronal nets decrease membrane capacitance of peritumoral fast spiking interneurons in a model of epilepsy. Nat Commun. 2018;9:4724. https://doi.org/10.1038/ s41467-018-07113-0. 11. Regalia G, Onorati F, Lai M, Caborni C, Picard RW. Multimodal wrist-worn devices for seizure detection and advancing research: focus on the Empatica wristbands. Epilepsy Res. 2019;153:79–82. https://doi.org/10.1016/j.eplepsyres.2019.02.007. 12. Han Z, et al. Antisense oligonucleotides increase Scn1a expression and reduce seizures and SUDEP incidence in a mouse model of Dravet syndrome. Sci Transl Med. 2020;12. https://doi. org/10.1126/scitranslmed.aaz6100.
77
Clinical 1. Spencer SS, et al. Predicting long-term seizure outcome after resective epilepsy surgery: the multicenter study. Neurology. 2005;65(6):912–918. 2. de Groot M, et al. Synaptic vesicle protein 2A predicts response to levetiracetam in patients with glioma. Neurology. 2011;77(6):532–539. 3. Allen AS, et al. De novo mutations in epileptic encephalopathies. Nature. 2013;501(7466):217–221. 4. Berg AT, Rychlik K, Levy SR, Testa FM. Complete remission of childhood-onset epilepsy: stability and prediction over two decades. Brain. 2014;137(12):3213–3222. 5. Devinsky O, et al. Cannabidiol in patients with treatment- resistant epilepsy: an open-label interventional trial. Lancet Neurol. 2016;15:270–278. https://doi.org/10.1016/ s1474-4422(15)00379-8. 6. Lambrechts DA, et al. A randomized controlled trial of the ketogenic diet in refractory childhood epilepsy. Acta Neurol Scand. 2017;135:231–239. https://doi.org/10.1111/ane.12592. 7. Cukiert A, Cukiert CM, Burattini JA, Mariani PP, Bezerra DF. Seizure outcome after hippocampal deep brain stimulation in patients with refractory temporal lobe epilepsy: a prospective, controlled, randomized, double-blind study. Epilepsia. 2017;58:1728–1733. https://doi.org/10.1111/epi.13860. 8. Sprengers M, Vonck K, Carrette E, Marson AG, Boon P. Deep brain and cortical stimulation for epilepsy. Cochrane Database Syst Rev. 2017;7:Cd008497. https://doi.org/10.1002/14651858.CD008497. pub3. 9. Thiele EA, et al. Cannabidiol in patients with seizures associated with Lennox-Gastaut syndrome (GWPCARE4): a randomised, double-blind, placebo-controlled phase 3 trial. Lancet. 2018;391:1085–1096. https://doi.org/10.1016/ s0140-6736(18)30136-3. 10. Lagae L, et al. Fenfluramine hydrochloride for the treatment of seizures in Dravet syndrome: a randomised, double-blind, placebo-controlled trial. Lancet. 2019;394:2243–2254. https://doi. org/10.1016/s0140-6736(19)32500-0.
I. STATIC NERVOUS SYSTEM DISEASES
This page intentionally left blank
S E C T I O N
I I
PROGRESSIVE NEURODEGENERATIVE DISEASES
This page intentionally left blank
C H A P T E R
4 Aging, Dementia, and Alzheimer Disease* Harald Sontheimer O U T L I N E 1
Case Story
81
2 History
82
3
Clinical Presentation/Diagnosis/Epidemiology 3.1 Epidemiology 3.2 Patient Presentation and Diagnosis 3.3 Alzheimer Disease 3.4 Frontotemporal Dementia 3.5 Other Causes of Dementia
83 83 84 85 86 87
6
Experimental Approaches/Clinical Trials
102
7
Challenges and Opportunities
104
Acknowledgments
105
Disease Mechanism/Cause/Basic Science 4.1 Memory and Distribution in the Brain 4.2 Structural Changes Underlying Alzheimer Disease and Other Forms of Dementia 4.3 Vascular Changes in Alzheimer Disease 4.4 The Genetics of Alzheimer Disease 4.5 Animal Models of Alzheimer Disease 4.6 Changes in Transmitter Systems in Alzheimer Disease 4.7 Alzheimer Disease and Inflammation
87 87
References
106
General Readings Used as Source
106
Suggested Papers or Journal Club Assignments Clinical Paper Basic Science Paper
106 106 107
4
5
89 95 95 96
Treatment/Standard of Care/Clinical Management 5.1 Imaging, Diagnosis, and Disease Prediction 5.2 Drug Treatment 5.3 Disease Onset Delay and Prevention
97 97 98 99
96 96
1 CASE STORY
i n-home care? I needed to work to keep up with bills and for my own sanity. I feel fortunate that the elementary school had an opening for me. Teaching distracts me from the daily guilt I feel for abandoning the wonderful woman who raised me. Mother is sitting in a reclining chair. Alicia, her nurse, is holding her hand while reading her a story. “Hello, Mom,” I announce myself. She looks up, bewildered. She has not spoken my name in many months. Slowly, a slight smile develops in her face, suggesting that maybe,
It is Easter Sunday. The lilies are finally breaking through the ground, which just weeks earlier bore a solid snow cover. I am packing the car to make the drive upstate to visit mother at Saddlebrook Plantation home. I am sad and angry at the same time, having moved her into assisted living just 9 months ago. But caring for her around the clock had become impossible. Should I have tried to keep her at home longer? Should I have hired
* The term “Alzheimer Disease” is often incorrectly used as possessive noun with an apostrophe (“Alzheimer’s disease”) in scientific and popular texts alike. This is likewise the case for Parkinson and Huntington disease. However, neither Alois Alzheimer, nor James Parkinson or George Huntington suffered from (or possessed) the respective illnesses named after them. Instead they are simply honored as having been the first to describe them. This is different from “Louis Gehrig’s disease” an alternative name used for Amyotrophic Lateral sclerosis or ALS, which became famous as the American baseball player Lou Gehrig actually suffered and died from this illness.
81
Diseases of the Nervous System. https://doi.org/10.1016/B978-0-12-821228-8.00004-4Copyright © 2021 Elsevier Inc. All rights reserved.
82
4. AGING, DEMENTIA, AND ALZHEIMER DISEASE
just maybe, somewhere deep in her brain, a memory of me remains. She slowly rises out of her chair and walks toward me. We hug, I hold her tight. She has become thin and frail, weighing barely 90 pounds. She reaches for the flowers I brought and slowly walks to retrieve a vase. She takes my hand and leads me to a craft table in her room with a number of small cardboard boxes spread out evenly. Some are colored; others are adorned with ribbons and bows. “Still need color…pretty.” She is searching for words. “These are pretty, Mom!” She looks up, her eyes fixated on the flowers. “Flowers?” Yes, Mom, I brought you flowers. Her hand shakes as she points and I am worried she may develop Parkinson’s as well. Isn’t she burdened enough? Mother is wearing at least three layers of sweaters although her room feels hot, almost uncomfortably so. “Let’s go visit your friends!” She walks slowly, holding on to my arm. We wander past the common area where a group of residents is playing board games. Others sit in wheelchairs, hunched over. Some are napping. I tell her about her grandchildren, Sam and Susi. She seems to listen without saying a word. I do not believe she remembers them. “Cuddles?” she asks. “Mom, Cuddles died 25 years ago. We have not had a dog since.” We walk in silence, taking comfort in each other’s company. She tires quickly, and she walks unsteadily. “I brought you a wheelchair,” I tell her. “No, no.” She was noticeably agitated by the thought of being confined to a wheelchair. She slowly labors along, holding on to my arm. She sits down in her recliner. It is quiet. She closes her eyes and drifts away. Her birthday is coming up. She does not know it. My next visit is two weeks away. I will bring her a cake. Alicia steps into the room to deliver afternoon tea. As Mom awakes, she startles, clearly surprised about my presence. Her memory of the past hours had already faded. Once again, she doesn’t remember me. It pains me. “She has been doing well the past week, but growing frail. Did you bring the chair?” I get the chair from the car and leave it with Alicia outside the room. “Hold off until it is absolutely essential. She hates the thought of being in a wheelchair.” I sit with Mom for what seems to be hours, remembering better days. At times, I cannot hold off the tears. She still sleeps. I kiss her on the cheek and steal myself out of the room. Sad, very sad, seeing her confined and robbed of all her past and present and unable to help. I still have not accepted her disease. I drive home to hug my two children and my husband, afraid that sooner or later I may suffer the same fate.
2 HISTORY Throughout the ages, the elderly grew forgetful, yet this cognitive decline was called “senile dementia” and considered a normal part of aging rather than a d isease.
FIGURE 1 Auguste D., the patient that Alzheimer described in 1906 and who served as the name-defining case.
Given that people rarely lived to old age, actual dementia was much less common. When it did occur at an early age, it was presumed to be the result of infections such as syphilis. In 1906 the German psychiatrist Alois Alzheimer described a case of severe dementia in a 51-year-old patient, Auguste D. (Figure 1), who was suffering from severe language deficits, delusions, hallucinations, paranoia, and aggression. She had been admitted to the state asylum in Frankfurt, Germany, where she remained Alzheimer’s patient until her death in 1906. Being a trained pathologist, Alzheimer performed an autopsy to examine her brain and found extensive atrophy (shrinkage) of the cortical gray matter not observed in normal individuals at that age. Taking advantage of new staining techniques to visualize microscopic cellular changes, he described abnormal neural fibrillary bundles and plaque-like extracellular deposits, the histopathological hallmarks that have become the defining features of Alzheimer disease (AD). In his original description, he wrote that “scattered throughout the entire cortex…one found military foci that were caused by the deposition of a peculiar substance…” that would become known as amyloid in 1984.1 For many years, this condition was considered an exceptionally rare form of early-onset dementia that became known as Alzheimer disease (AD) through a psychiatry textbook published by Alzheimer’s mentor Emil Kraepelin in 1910. Not until 1976 was AD broadly considered to cause senile dementia in the elderly, when Robert Katzmann examined countless autopsies from patients across the age spectrum and found that the pathological changes described by Alzheimer were indeed a common feature in all of them. The physical nature of plaques and tangles remained a mystery until the 1980s when both amyloid and tau were identified as the principal molecules constituting
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
3 Clinical Presentation/Diagnosis/Epidemiology
plaques and tangles, respectively. Since then, numerous studies have examined the relative importance of neurofibrillary tangles and amyloid deposits as disease- causing agents in AD, leading to the current view that abnormal amyloid accumulation is the most likely primary cause of the progressive cognitive decline in AD. This hypothesis is supported by genetic studies of rare familial forms of AD. The initial evidence that AD may have a genetic component came from a study that examined the relatives of 125 individuals with autopsy-confirmed AD. These families had an unusual increase in the number of individuals with dementia, and also a surprisingly greater incidence of Down Syndrome. This mysterious connection to Down Syndrome, trisomy 21, became clear as the first disease-causing mutation in a Dutch family mapped to chromosome 21. In turn, this led to the identification of the amyloid precursor protein (APP) that gives rise to the disease-causing toxic forms of β-amyloid. Individuals with Down Syndrome carry an extra copy of the APP gene, explaining the relatively common occurrence of dementia in Down Syndrome. Since the positional cloning of the APP in 1987, over 50 mutations have been identified in the APP gene and provide the framework for our current genetic understanding of AD.2 Additional mutations in the presenilin 1 gene were discovered in 1995 and the ε4 allele of the apolipoprotein (APOE) gene was discovered as a major risk factor for AD in 1997. Although AD is only one of many causes of dementia, it vastly outnumbers all other causes and consequently has become synonymous with dementia in the popular press.
3 CLINICAL PRESENTATION/ DIAGNOSIS/EPIDEMIOLOGY Aging is a normal and natural process. Although it carries a negative connotation, this is not warranted, since healthy, successful aging is a positive life experience that is typically associated with favorable changes in judgment and personality. For example, the ability to make thoughtful decisions that are not overtly influenced by the immediate emotional state is a skill acquired with age and experience. Similarly, the ability to solve seemingly overwhelming problems and place them into proper context improves with age. In most individuals, aging is accompanied by a measurable but tolerable loss in cognitive ability that does not significantly affect a person’s quality of life or ability to function. As the brain ages, opportunities for subtle neuronal changes compound over time and can give rise to a pathological decrease in mental capacity. Neurologically, older individuals tend to be a little more forgetful. Misplaced car keys and wallets and difficulty remembering names are typical examples that frighten people to consider the possibility that they may suffer from a pathological loss
83
of memory. Typically, none of these are of clinical concern, but all are the source of many jokes. Clinically, this is labeled “benign forgetfulness of the elderly.” If memory loss progresses further, and is associated with signs of personality changes, fatigue, restlessness, and irritability, these signify unhealthy brain aging and may be early symptoms of dementia. Dementia is a serious loss of global cognitive ability in a previously unimpaired individual and is significantly beyond what would be expected from normal aging. Dementia is a syndrome rather than a disease, as it can arise from numerous different causes. Some dementias are static, most notably if the underlying cause was trauma or a stroke; some are reversible, as is the case with drug abuse, where dementias resolve after cessation of drug use. The majority of dementias, however, are progressive and are typically associated with neurodegenerative diseases. The most important forms of neurodegenerative diseases associated with dementia include Creutzfeldt-Jakob disease (CJD), dementia with Lewy bodies (DLB), frontotemporal dementia (FTD), and AD. As elaborated further below, these diseases typically present with abnormal aggregations of specific proteins; for example, Aβ42 and tau in AD, α-synuclein in DLB, tau or TDP-43 in FTD, and misfolded prion protein (PrPsc) in CJD. Among the elderly, AD is by far the most common cause of dementia, and we will devote most of this chapter to it. Note, however, that FTD is the largest cause of dementias in people under 65, where AD is uncommon. Vascular disease, stroke, trauma, and brain tumors or infections with human immunodeficiency virus (HIV) or syphilis are additional causes of dementia. More rare causes include vitamin B1 or B12 deficiencies.
3.1 Epidemiology Worldwide, over 43 million people are affected by dementia, with a new case developing every 3 s. Frighteningly, this number has more than doubled in the past 25 years. As graphically illustrated in Figure 2, the disease burden is disproportionately high in the industrialized world, in large part due to the significantly longer life expectancy afforded by superior nutrition and health care. Across all age groups, dementia is the 5th leading cause of death globally, but is second overall if one considers only people over 70 years who are most likely affected by dementia. AD is by far the leading cause for dementia and currently affects 34 million people worldwide and 5.7 million in the United States. These numbers are expected to double by 2030 and quadruple by 2050 with the potential to overwhelm our healthcare system. Treating and caring for individuals with dementia already consumes over $600 billion worldwide. African Americans have a 2.5-fold increased risk compared to Caucasians, possibly due to increased rates of
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
84
4. AGING, DEMENTIA, AND ALZHEIMER DISEASE
FIGURE 2 Worldwide disease burden of dementias, expressed as the disability-adjusted life year (DALY), which shows the number of years lost due to ill health, disability, or early death per 100,000 inhabitants in 2004. Wiki Commons, based on 2004 World Health Organization Data.
diabetes, which alone increases the risk for AD by threefolds. The Framingham study followed nearly 2800 people who at age 65 were healthy and free of dementia over a 29-year period and documented the development of dementia. It concluded that, on average, a 65-year-old woman has a 21.7% chance of developing dementia and a 17.2% chance of developing AD in her lifetime, whereas a 65-year-old man has a 14.3% lifetime risk for dementia and a 9.1% chance for AD.3 Women are almost twice more likely to die from dementia than men and this is only partially explained by women living longer. Numerous environmental risk factors have been proposed to cause AD, but none have been confirmed as actual risk factors to date. By far the greatest known risk factor for all dementias, including AD, is age, with an exponential increase in risk as a function of age. Some scientists have speculated that dementia is an inevitable consequence of aging. However, many centenarians (individuals over 100 years of age) have normal memory function. A family history of dementia increases the risk to develop AD, with first-degree relatives having a twofold increased risk of developing AD. Expression of the apolipoprotein E4 (APOE-ε4) is significantly correlated with late-onset AD. The risk of AD increases up to 15fold if both alleles are affected compared to a fourfold if only one ε4 allele is present and the patient has a family history of AD. Smaller increases in risk ranging from 1.5- to 1.9-fold are attributable to modifiable risk factors such as hypertension, obesity, diabetes, inactivity, social isolation, smoking, and hearing loss in midlife. In an effort to stem this epidemic, many studies have been conducted to explore whether dietary supplements,
nutrition, or lifestyle may reduce the likelihood to develop dementia. Thus far, the data are incomplete or inconclusive. Some credible studies suggest that the Mediterranean diet may reduce the risk of AD,4 and individuals with higher levels of education are less likely to develop AD. This may be due to a greater “cognitive reserve” capacity, although one must consider that education often correlates with income, nutrition, and access to health care, all of which may directly or indirectly affect the likelihood to develop disease. Interestingly, recent imaging studies found that cognitive engagement in midlife slows the build-up of toxic β-amyloid deposits,5 the pathological hallmark for AD, and suggest that brain activity may indeed modify disease etiology.
3.2 Patient Presentation and Diagnosis The salient deficit that characterizes all forms of dementia is impaired memory and cognitive function. Unfortunately, we currently have no biomarkers or noninvasive tests to unequivocally diagnose an individual with dementia. Hence the initial diagnosis of suspected dementia relies on cognitive tests, family history, and a comprehensive neurological examination. Whenever memory deficits are suspected, memory tests should be conducted by a neurologist or neuropsychologist. The easiest way to do so in the clinic or at the bedside involves the Mini-Mental Status Examination (Table 1), which is a 30-point test of both working and episodic memory. Tasks include knowledge of the patient’s current whereabouts, naming objects, and recalling a series of objects and words. An easy way to test
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
3 Clinical Presentation/Diagnosis/Epidemiology
85
TABLE 1 Memory assessment using the Mini-Mental Status Exam. Points Orientation Name: Season/date/day/month/year
5 (1 for each name)
Name: Hospital/floor/town/state/country
5 (1 for each name)
Registration Identify three objects by name and ask patient to repeat
3 (1 for each object)
Attention and calculation Subtract serial 7s from 100 (93-86-79-72-65)
5 (1 for each subtraction)
Recall Recall the three objects presented earlier
3 (1 for each object)
Language Name pencil and watch
2 (1 for each object)
Repeat: “No ifs, ands, or buts”
1
Follow three-step command (take paper/fold/place on table)
3 (1 for each command)
Write “close your eyes” and have patient follow written command)
1
Ask patient to write a sentence
1
Ask patient to copy a design (pentagon, triangle, etc.)
1
Total
30
The mini-Mental Status Exam allows for a rapid assessment of a person’s memory mental and memory function. A total of 30 points are available for correctly answering 30 questions in 5 categories; 27 or more points are considered normal; 19–24 points are indicative of mild cognitive impairment; 10–18 moderate and fewer than 10 points suggest severe cognitive impairment.
working memory is to ask a patient to repeat a series of numbers either forward or backward. Most normal adults can repeat six digits forward and five digits backward. Early signs of forgetfulness must be calibrated to age, recognizing that an 85-year-old remembers only half as many words in a list as an 18-year-old. The smallest measurable deviation in memory function derived from memory tests is called mild cognitive impairment (MCI), which is a measurable deficit in memory, but one that does not interfere with daily living. Individuals who fall in this category typically score at least a 23 or higher in the mini-mental exam. MCI may be a precursor to AD and may progress to disease, particularly in patients who bear an increased risk due to family history of AD or carrying the APOE-ε4 allele, discussed in greater detail later.
3.3 Alzheimer Disease Alzheimer Disease is often used synonymously with dementia because it is the single most prominent cause of dementia, accounting for ∼ 80% of all dementias worldwide. Ten percent of all individuals aged over 70 years have mild memory impairments, but half of these will progress to AD within 4 years. Three-quarters of all AD patients initially present with memory problems, including misplacing objects, being unable to manage money, getting lost, and not following instructions. If these
roblems progress slowly, the patient is likely to have p AD. Motor functions are typically spared until much later in the disease. Mild depression and social withdrawal are common. Unfortunately, most clinical tests, except for amyloid imaging, which is used experimentally, can only rule out other conditions, such as vitamin deficiencies, but cannot unequivocally diagnose AD. As we will discuss later, this may soon change through recently developed noninvasive ways to measure amyloid deposits in the brain. However, currently, the pathological hallmarks of AD can only be assessed on autopsy after a patient’s death or on imaging for amyloid (discussed later). AD presents with a similar pattern in most patients and is often divided into clinical stages. During the early stage, lasting 2–4 years, patients have frequent memory losses affecting recent memories but preserving old memories. For example, patients may forget entire conversations, cannot follow directions, and routinely forget where they have placed things, yet remember the names of family members, friends, and relatives. Although conversational language is initially seemingly intact, language is less fluid and language comprehension declines gradually. A person with AD may place objects in odd places, for example, leaving his or her keys in the refrigerator. Mood swings occur frequently and patients may be apathetic, with signs of depression. They may need to be motivated and told what to do on a daily basis.
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
86
4. AGING, DEMENTIA, AND ALZHEIMER DISEASE
In the second stage of the disease, lasting between 2 and 10 years, patients are no longer able to hide their handicaps and problems. Memory loss becomes much more pervasive and begins to involve older memories as well. Patients may no longer remember the names of their friends and family. Patients may claim that a caregiver is an imposter (Capgras syndrome). Their speech is often nonsensical. They become easily disoriented and get lost in formerly familiar places, and may no longer find their way home if left alone. Patients withdraw and isolate themselves in social situations. They forget social behavioral norms, and a loss of inhibition can lead to awkward behaviors. Memory problems severely interfere with daily activities, such as driving, shopping, or balancing accounts. Patients show difficulties dressing, eating, and even walking. Patients look and walk “Parkinsonian,” and are stiff and rigid. At the final stage of disease, lasting from 1 to 3 years, patients are confused and generally unable to carry out any conversation. They have major mobility difficulties. Problems swallowing increase the risk of aspiration (lodging food or drink in the airways) and are also the most common cause of death in AD. During this stage of the disease, individuals require constant supervision, often in a skilled nursing facility. Overall disease duration ranges from 1 to 25 years, but the typical life expectancy from disease onset ranges between 8 and 10 years. The constant care required places a tremendous burden on family and caregivers, and may isolate spouses or family members, who themselves are at risk to develop apathy and depression due to the stress of being on call around the clock.
3.4 Frontotemporal Dementia FTD is the most common dementia diagnosis in individuals less than 65 years of age, and disease onset
can be as early as age 45. In the United States, approximately 50,000 individuals currently suffer from FTD with approximately 10,000 new cases diagnosed each year. Although FTD is about equally as common as AD in the population under 60, AD is far more common (100-fold) across all age groups. The deterioration of the frontal lobe (Figure 3) causes changes in social and personal behavior. Apathy is a common signature, leading to social withdrawal, with patients frequently staying in bed all day. A lack of inhibition leads to neglect of social norms and to inappropriate behaviors that may include stealing, speeding, and disinhibited sexual drives. Compulsive behaviors develop along with binge eating. As disease progresses, patients lose fluency in their speech, primarily because they are unable to articulate language, yet they are without loss of language comprehension. The loss of executive function compromises daily planning and organization. Taken together, the most defining clinical features of FTD include disinhibition, apathy, loss of empathy, compulsive behavior, hyperorality, and loss of executive function. Structural magnetic resonance imaging (MRI) often reveals the selective atrophy of the frontal lobe, easily distinguishing the disease from AD. The same differentiation can also be seen on fluorodeoxyglucose positron emission tomography (FDG-PET) (Figure 15), where glucose utilization is selectively reduced in the frontal lobe in FTD, whereas in AD, the decrease in glucose utilization is more posterior. A number of genetic mutations have been identified and include the gene encoding for the microtubule-associated tau protein. There is no known cure and treatment aims at management of behavioral symptoms. Patients typically die within 2–10 years after diagnosis, with continuous supervision required in end stages of disease. Given its early onset, often in the prime productive years of a person’s life, the socioeconomic effects are particularly devastating.
FIGURE 3 Two brains obtained on autopsy comparing an individual that suffered from frontotemporal dementia (FTD) to that of an age matched a normal person. The arrows indicate the selective shrinkage (atrophy) of the frontal loves in FTD. Courtesy of Dr Peter Anderson, University of Alabama Birmingham, Department of Pathology & Pathology Education Informational Resource (PEIR) Digital Library (http://peir.path.uab.edu).
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
4 Disease Mechanism/Cause/Basic Science
3.5 Other Causes of Dementia Numerous other conditions can cause dementia, including vascular disease, tumors, brain trauma, and viral and bacterial infection, as well as drug or alcohol abuse. Many of these conditions produce a static disease that may be reversible if the initiating insult is removed. For example, drug- induced dementia may disappear upon drug withdrawal. Acquired immunodeficiency syndrome (AIDS)-associated dementia has gained increasing importance, since patients infected with the HIV virus now live an almost normal life span, yet frequently develop dementia as a consequence of viral damage to the brain. This condition, discussed more extensively in Chapter 10, is typically associated with low CD4 T-cell counts and widespread brain inflammation. In the developed world, the widespread use of highly effective antiviral therapy has made AIDS-associated dementia less common, yet this form of dementia still poses a significant challenge in the developing world.
4 DISEASE MECHANISM/CAUSE/BASIC SCIENCE From a functional point of view, the changes that most dramatically distinguish unhealthy aging and disease from normal healthy aging are altered memory function, personality changes, and compromised executive function. Central to these impairments are the loss of short- and long-term memory and an inability to form new memories. Structural changes, often only accessible in autopsy tissue, include vascular changes with reduced cortical blood flow and breaches in the blood–brain barrier (BBB), extracellular amyloid deposits, intracellular
87
accumulation of neurofibrillary tangles, reduced neuronal numbers, reactive gliosis, and microgliosis. We will extensively discuss each of these below. Since loss of memory is the functional hallmark of all dementias, it is useful to briefly review the various forms of memory and their anatomical and physiological substrates, as this helps to explain the pattern by which lesions in the brain selectively rob certain memories while sparing others.
4.1 Memory and Distribution in the Brain Based on the duration with which memories are being retained, we typically distinguish short-term and longterm memory, also called proximal and distal memories. As the name implies, short-term memory is fleeting, lasting only seconds to minutes, and is highly sensitive to distractions, whereas long-term memories, by contrast, can last up to a lifetime. Working memory has been suggested as a more suitable substitute for short-term memory, since it captures the concept that those memories require attention and are subject to alteration. A sensitive test for working memory requires a subject to spell a word backward or to repeat a series of numbers in reverse order. These tasks are not a matter of simple recall; rather, they require significant processing. Imaging studies and evaluation of patients suffering from brain lesions each suggests that the prefrontal cortex plays an important role in working memory. Working memory declines with age and is one of the early deficits in AD. Our long-term memory is often divided into declarative memory, which is the ability to remember facts, events, and names, and nondeclarative memory, which are skills that we perform without awareness, for example, riding a bicycle (Figure 4). As we will soon see, these
FIGURE 4 Forms of memories. Declarative memories involve the hippocampus and are often termed the “where” memories. Nondeclarative or procedural memories are the “how” memories, and do not involve the hippocampus.
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
88
4. AGING, DEMENTIA, AND ALZHEIMER DISEASE
r eside in separate areas of the brain and, accordingly, are differentially affected in AD. Declarative memories that capture personal experiences, such as a fishing trip with your dad, a recent birthday, or your high school graduation, are examples of episodic memory. Memories that encompass general knowledge, such as the use of language, the number system, the purpose and use of tools, classes of animals and plants, and the like, are called semantic memories. Semantic memory is typically retained for a very long time, as are most episodic memories, although some are kept only for a few minutes or hours and then discarded. Both semantic and episodic memories are required in tandem to generate stories with narrative, relating “what” happened to “whom,” “when,” and “where.” Declarative memories are believed to be stored in the cortical regions where the experiences associated with these memories, such as sounds, smells, and faces, were perceived and processed at the time. Semantic memories are associated with the brain region responsible for the particular semantic task. For example, the concept of language and numbers is stored in the parietal cortex associated with speech and language, while the use and concept of tools resides in the motor cortex. Recognition of colors, faces, objects, animals, and the like are stored in the somatosensory cortex. Spatial memories reside in the posterior hippocampus, making the hippocampus essential for spatial navigation. Taken together, various individual facts underlying a story, including its content, faces, locations, smells, and sounds, are all stored in a distributed cortical and hippocampal network. To recall an experience, this information needs to be brought together in an organized train of thought or narrative. This occurs through the cortical–medial temporal lobe (MTL) network illustrated in Figure 5, which encompasses the hippocampus, entorhinal cortex, and the perirhinal cortex with its bidirectional projections to the cortex. The cortical memory traces are retrieved via the entorhinal cortex and modulated within the circular projections of the entorhinal cortex to the dentate gyrus, to the CA1 and CA3 neurons of the hippocampus, and back to the entorhinal cortex. With every passage of information through the cortex–MTL network, memories are strengthened or weakened, presumably within the circuitry of the hippocampus. The more often such a memory is recalled, the stronger and more long-lasting it becomes. It is important to point out that the cortex–MTL pathway is constantly modulated by input from the amygdala, providing an emotional input to our memories. Not surprisingly, we tend to form longer-lasting memories more quickly if they occur in a state of either intense fear or intense pleasure. Interestingly, this memory pathway is involved not only in recalling past memories but also in envisioning future events. Functional MRI studies show
FIGURE 5 Declarative memories are distributed throughout the cortical–mediotemporal lobe network and require information flow through the hippocampus and the surrounding entorhinal and perirhinal cortices.
activity in the hippocampus when individuals are asked to project into the future, and persons with bilateral hippocampal lesions are unable to visualize future events. Loss of the hippocampus and its associated entorhinal and perirhinal cortices causes anterograde amnesia, or the inability to form new memories, yet it preserves very old memories (> 20 years). Damage to any part of the cortical–MTL circuit presents with memory loss, and AD affects essentially all aspects of the cortical–MTL memory pathway at some point in the disease. Indeed, the sequential development of lesions beginning in the hippocampus and surrounding entorhinal cortex, gradually spreading into the cortical gray matter, illustrated in Figure 6, explains the sequential loss of short-term followed by long term episodic memory. The importance of the hippocampus for the acquisition of new memories is most effectively demonstrated by the famous story of H.M., who had both hippocampi surgically removed to treat his intractable epilepsy. This left him unable to remember anything at all for longer than a few minutes, yet he was able to recall old memories acquired prior to his surgery, and maintained intact procedural memories. The early loss of the hippocampus in AD explains why patients have difficulty acquiring new memories early in the disease process, while sparing recall of old memories. It also underlies the early deficits in spatial memory stored in the hippocampus,
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
4 Disease Mechanism/Cause/Basic Science
89
FIGURE 6 Schematic depiction of the spread of AD throughout the brain. During preclinical phases the primary affected areas are the hippocampus and entorhinal cortex. This pathology would sometimes be labeled Braak stage 1 or 2. Pathology spreads to include the frontal lobe, at which point it would be categorized Braak stage 3 or 4. Ultimately the entire cortex and even cerebellum present with wide-spread pathology. This would represent a Braak stage of 6. Adapted from a picture from National Institute for Aging (NIA).
which explains the frequent symptoms of getting lost and not remembering where they left objects. Nondeclarative memory, or the ability to perform tasks in a nonconscious way, such as using silverware to eat, riding a bicycle, or playing an instrument, is anatomically and functionally entirely different from other forms of memory. This form of memory can also be called procedural memory and is typically preserved even if declarative memory is lost. In other words, you may not remember practicing or taking piano lessons, but can still be able to play the instrument. Examples of procedural memories are playing sports and riding a bicycle, which typically take a long time to acquire but stay with you for life. Grandma still remembers how to ride a bicycle even if she has not used one in ages and may no longer remember where she stored it. It is the “knowing how” of nondeclarative memories that separates them from the “knowing that,” which characterizes declarative memories. Nondeclarative memories are not hippocampal-dependent, and instead involve the basal ganglia, amygdala, cerebellum, and sensory and motor cortices, and are locally stored in those areas that are normally part of our motor system. The hand and finger representations of musicians occupy larger areas on the
somatosensory cortex than in nonmusicians. Motor procedural memory tasks are stored in the basal ganglia and cerebellum and are therefore, not unexpectedly, compromised in Parkinson Disease but are largely intact in AD, which spares these structures. Memory at the cellular level: To understand the pathological changes that occur to memory function in aging, dementia, and AD, it is also important to briefly review our knowledge of learning at the cellular level. Two important synaptic changes that are called longterm potentiation (LTP) and long-term depression (LTD) are now widely accepted as representing the cellular and neurochemical basis for learning and forgetting, respectively. These are briefly reviewed in Box 1.
4.2 Structural Changes Underlying Alzheimer Disease and Other Forms of Dementia Anatomical Changes In AD, pathological changes begin in the transentorhinal region, spreading to the hippocampus, and then move to the lateral and posterior temporal and parietal neocortex (Figure 6). Eventually AD presents with widespread and more diffuse degeneration and a global
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
BOX 1
N E U R O C H E M I C A L B A S I S O F M E M O R Y ( LT P A N D LT D ) Our cellular and molecular understanding of memory is very incomplete and is largely based on the study of model systems. Vertebrate memory studies have largely focused on the hippocampus, where a commonly used conditional learning paradigm involves stimulating a group of neurons, followed by a second train of stimulation shortly thereafter. The second train causes the postsynaptic neurons to show a much larger response that persists for minutes to hours. This learning of coincident activity is called long-term potentiation (LTP) and involves both the NMDA-type Glu receptor, which is designed to act as a coincidence detector, and the AMPA receptor, which mediates fast synaptic transmission. A single action potential arriving at a presynaptic terminal would give rise to only a short postsynaptic response mediated by the AMPA-R, as the NMDA-R is blocked by intracellular Mg2 + ions. If the cell receives a strong input, either through repeated activation from the same synapse or from some of its other dendrites, the cell depolarizes just enough to allow an Mg2 + ion to dislodge from the pore of the NMDA-R, unblocking it. Any subsequent stimulus arriving within about 50 ms will permit the influx of Ca2 + into the postsynaptic cell through the now unblocked NMDA-R. This increase in Ca2 + triggers long-term changes in the postsynaptic cell involving activation of calcium-dependent protein kinase (CaMK). This leads to increases in the number of AMPA-Rs inserted into the membrane, making the response to subsequent stimuli larger. Central to the idea that this process is a cellular form of memory is the fact
that the synchronous activity of two neurons that fire together enhances the response between them and, over time, strengthens their connectivity, while cells that are asynchronous in their activity show no enhancement. Just as LTP enhances the connection between cells, and therefore “learns” a connection, an opposite cellular equivalent of forgetting exists as well. This is called long-term depression or LTD. Here, long-term stimulation of one cell at a low frequency, about once per second over 10–15 min, reduces the postsynaptic activity through a reduction in the number of functional AMPA-Rs. Whether LTP or LTD occurs depends on the size of the Ca2 + change in the postsynaptic terminal. Small, long-lasting changes in Ca2 + cause LTD; brief, large changes cause LTP. LTP and LTD are still short-term associative learning phenomena that result in relatively short-lasting (minutes to hours) changes in the physiology and are the result of phosphorylation of AMPA-R that changes their trafficking into and out of the membrane. These can, however, lead to long-lasting memory traces that require changes in the biosynthesis of these receptors, and probably changes in the neural network. For purposes of AD, it is important to remember that both NMDA- and AMPA-type Glu receptors, particularly those in the hippocampus and entorhinal cortex, are essential for learning and memory, and that NMDA-Rs are the target of memantine, a use-dependent inhibitor of NMDA and one of the few drugs approved to treat cognitive decline in AD. BOX 1-FIGURE 1 Neurochemical basis of memory (LTP and
LTD). High-frequency stimulation dislodges Mg2 + from NMDA receptors, allowing enhanced influx of Ca2 +, which leads to the insertion of AMPA receptors into the postsynaptic membrane, causing an increase in postsynaptic current (LTP). Low-frequency stimulation, on the other hand, leads to much lower postsynaptic Ca2 + and removal of postsynaptic AMPA receptors, leading to a reduced postsynaptic current (LTD).
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
4 Disease Mechanism/Cause/Basic Science
thinning of the cortical gray matter, readily visible across the entire brain on autopsy (Figure 7). The time course of anatomic lesions explains the gradual appearance of symptoms, from early memory deficits to aphasia (loss of language ability) and later navigational issues. This pattern of functional loss distinguishes AD from other dementias, most notably, FTD (Figure 3), where selective lesions in the frontal lobe cause problems in executive function, such as loss of judgment, as initial symptoms. Upon autopsy, two characteristic abnormalities can be observed that have become the hallmarks for AD and other dementias. These are neurofibrillary tangles and amyloid plaques (Figure 8). Their gradual appearance in the affected brain regions has been suggested to be either the cause of or consequence of neuronal death. Pathologists grade the level and location of pathology using six stages named Braak stages. Stage 1 signifies infrequent pathology restricted to the entorhinal cortex (Figure 6, Preclinical AD), Stage 3 involves limbic system structures and hippocampus (Figure 6, Mild to Moderate
91
AD), and Stage 6 represents the most severe pathology with widespread tau and amyloid pathology throughout the cerebral cortex (Figure 6, Severe AD). Neurofibrillary Tangles
Neurofibrillary tangles (NFTs) are always present in AD autopsy specimens. They are entirely made up of the microtubule-associated protein tau, which, when hyperphosphorylated, forms insoluble aggregates that can fill the entire intracellular space of a neuron.7 Six isoforms of tau derive from a single gene on chromosome 17. The longest tau isoform contains 441 amino acids and 79 possible phosphorylation sites. In normal neurons, only three residues are phosphorylated, as this promotes tubulin assembly into microtubules. Through its interaction with microtubules, tau serves both a structural and dynamic role. Microtubules run the length of the axon, which can be up to 1 m long, and are an important cytoskeletal element. In addition, microtubules serve as tracks along which cargo is moved from the cell body to the axon terminal. Such cargo includes proteins, lipids,
FIGURE 7 Autopsy shows extensive brain atrophy in a patient who died from AD (right) compared to a similar-aged individual who died from natural causes (left). Images in (A) and (B) show overall brain atrophy, while (C) and (D) show marked thinning of the cortical gray matter and enlargement of the ventricles in the AD patient. Courtesy of Dr Peter Anderson, University of Alabama Birmingham, Department of Pathology & Pathology Education Informational Resource (PEIR) Digital Library (http://peir.path.uab.edu).
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
92
4. AGING, DEMENTIA, AND ALZHEIMER DISEASE
FIGURE 8 Neuropathological hallmarks of Alzheimer Disease, neuritic plaques, and neurofibrillary tangles visualized with modified Bielschowsky stain. These are from the autopsy of an 83-year-old woman with AD. Courtesy of Dr Steven L. Carroll, Medical University of South Carolina.
FIGURE 9 Schematized role of tau protein in healthy (left) and diseased (right) brain. Tau stabilizes the microtubules along which cargoes such as neurotransmitter vesicles are moved. The hyperphosphorylation of tau causes a loss of microtubule stability, disrupting transport of cargo vesicles. From National Institute for Aging (NIA).
synaptic vesicles, and even mitochondria (Figure 9). The motor proteins that move cargo along microtubules include kinesin and dynein. Some cargo molecules move fast, 50–400 mm/day; others move slowly, 8 mm/day. Hyperphosphorylation of tau prevents its binding to microtubules, which in turn become unstable and disintegrate, hence compromising both the axonal cytoskeleton and the tracks needed for axonal transport (Figure 9), which both may hasten neuronal cell death. In addition to the disruption of microtubules, hyperphosphorylated tau self-aggregates into insoluble inclusions, the NFTs, which no longer bind tubulin. Tau pathology is an early feature of AD, and appearance of tangles correlates with
neuronal loss; however, in animal models of AD, considerable neuronal death can precede NFTs, suggesting that NFTs may be a consequence of disease rather than a cause. The neurotoxicity of NFTs has also been called into question. Once assembled into NFTs, tau becomes resistant to proteolysis by calcium-activated proteases, and it has therefore been suggested that tau aggregation serves to protect neuronal tau in stressed neurons. Indeed, even in the presence of NFTs, neurons can survive for years. Although a hallmark for the diagnosis of AD, tau aggregates are also found in FTD and several “Parkinsonian” Diseases, such as progressive supranuclear palsy. These diseases are often called “tauopathies.”
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
4 Disease Mechanism/Cause/Basic Science
FIGURE 10 Sequential cleavage of amyloid precursor protein (APP)
by β- and γ-secretase produces oligomeric Ab40 and 42 fragments. N and C designate the N- and C-terminal of the APP protein. Note that γ-secretase cleaves within the membrane, whereas β-secretase cleaves on the extracellular plasma membrane site. Cleavage by α-secretase produces a soluble longer form of amyloid that is not toxic.
Amyloid
Plaque deposits can be found in the brains of patients with a variety of neurodegenerative diseases. These plaques are generally insoluble aggregates containing a number of misfolded proteins that occur naturally in the body. Being misfolded, these proteins stick to each other, forming hydrophobic fibrils. The specific plaques found in the brains of patients with AD contain an overabundance of beta amyloid (Aβ), a 36–43 amino acid peptide. It is generated from the amyloid precursor protein APP through a series of cleavages by enzymes called secretases. The APP is a large transmembrane glycoprotein of unknown function expressed in neurons and nonneuronal cells. As illustrated in schematic form in Figure 10, the bulk of APP is cleaved by the enzyme α-secretase, yielding a soluble APPα and, after further cleavage by γ-secretase, the intracellular domain produces a signaling domain that acts as transcriptional regulator. An alternative cleavage pathway involves the sequential cleavage by β- and γ-secretase, and produces soluble amyloids ranging from 36 to 43 amino acids in length, with the most common variants being Aβ40 and Aβ42. Plaques contain both Aβ40 and Aβ42, with Aβ40 being the more abundant of the two. However, Aβ42 is more hydrophobic and therefore the most fibrillogenic, and is considered the more toxic of the two. In neurons, unlike in nonneuronal cells, the production of Aβ40 and Aβ42 can also occur intracellularly,8 and intracellular Aβ can be released from neurons at synapses in an activity- dependent manner. Autosomal dominant mutations in the APP gene cause an increase in the relative production of the plaque-forming Aβ42 and characterize rare familiar forms of AD. Moreover, it is now widely accepted that abnormal expression or processing of APP causes an
93
imbalance between Aβ production and clearance, and that this causes the gradual build-up of Aβ, leading to the formation of plaques in both sporadic and familiar forms of AD. While the disease association of plaques and AD is generally accepted, it is less clear whether the plaques in and of themselves are neurotoxic and sufficient to explain the gradual cognitive decline in AD.9 Plaques are typically surrounded by soluble Aβ oligomers, which are in equilibrium with the plaques acting as storage sites for Aβ oligomers (Figure 11). These oligomers may play a major pathophysiological role in disease. Simply applying soluble Aβ oligomers on hippocampal slices impairs synaptic function, and specifically blocks LTP while enhancing LTD, the cellular substrates of learning and unlearning, respectively (see Box 1-Figure 1). Moreover, neuronal activity increases the production and release of soluble Aβ, which then acts as a feedback regulator on synaptic transmission by reducing neurotransmitter release. Pathologically elevated levels of Aβ would be expected to put this feedback loop into overdrive, depressing excitatory transmission. Some studies even suggest that the relative concentration of soluble Aβ42 at the synapse determines whether synapses are depressed or potentiated. These findings provide a potential normal physiological role for Aβ, acting as a feedback regulator at synapses, while also explaining the abnormal pathological role when too much Aβ causes profound changes in synaptic activity and build-up of plaques. The idea that soluble Aβ may cause aberrant synaptic activity would explain how cognitive changes and memory loss could present long before significant plaque burden is present. Precisely how Aβ regulates synaptic transmission in normal brain and in AD remains to be clarified, and the cellular receptor(s) for Aβ has (have) yet to be identified. Secretases As stated above, the default pathway for APP cleavage is via α-secretase, which releases a soluble APPα protein and, after further cleavage by γ-secretase, an intracellular transcriptional regulator P3. β-Secretases attack the APP protein at a different site, leaving a longer segment of APP membrane bound, which, upon cleavage by γ-secretase, forms the various toxic forms of Aβ. Our understanding of secretases is still emerging. This group of molecules is functionally defined by their proteolytic activity whereby they liberate a substrate from the membrane in a secreted form. γ-Secretase is not a single protein, but instead it is a complex of four proteins that include nicastrin, APH-1, PEN-2, and presenilin. The proteolytic activity is conferred by presenilin, yet the other three proteins are essential regula-
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
94
4. AGING, DEMENTIA, AND ALZHEIMER DISEASE
FIGURE 11 The dynamic relationship between plaques and oligomers with differential effects on brain physiology and pathophysiology. Oligomers may affect signaling functions of neurons, while plaques either act as storage sites or directly attract and activate microglia, eliciting an inflammatory response. Reproduced with permission from Ref. 10.
tors of APP cleavage. Presenilins are 8-transmembrane domain proteins that derive from two genes (PSEN1 and 2). Knockout of PSEN1 kills mice at an embryonic stage. This is because γ-secretases have other important roles, particularly during development. Most importantly, a transmembrane receptor called Notch that is expressed on the surface of many cells, including neurons, requires γ-secretase activity to signal. Notch mediates cell–cell signaling required for cell differentiation and overall establishment of polarity in a multicellular organism. Adjacent cells each express both the Notch receptor and a membrane anchored ligand, which in mammals is either Jagged or Delta-like. Upon binding to Notch, the intracellular domain of Notch is cleaved by γ-secretase and then trafficked to the cell nucleus to regulate gene transcription. As illustrated in Figure 12, γ-secretases serve a dual role, enabling Notch signaling for orderly development while also producing Aβ.12 While this dual role may have evolved by design, it is possible that only the Notch receptor was the intended substrate and that APP cleavage is an erroneous process that has the potential to be pathogenic. Notch signaling remains important in the adult nervous system, where it has been implicated in learning and memory, and the importance for γ-secretase in nor-
mal Notch signaling provides a challenge to clinical trials that target Aβ production, further discussed later. Clearance of Deposits In addition to overproduction of Aβ and change in the ratio of Aβ40/42, impaired Aβ clearance has emerged as a potential contributor to disease. Several clearance mechanisms for Aβ have been identified. For example, the low-density lipoprotein receptor-related protein (LRP) can form a complex with Aβ on the abluminal side of the blood vessel endothelial cells. These complexes are then either internalized and degraded by the lysosome or released by transcytosis across the BBB into the bloodstream. Clusters of Aβ can be visualized in association with blood vessels in live mice that harbor mutations in the hAPP gene (Figure 13). Other pathways for degradation involve proteolysis via plasminogen or insulin-degrading enzyme and phagocytosis by astrocytes and microglial cells. Interestingly, amyloid may be cleared through the cerebrospinal fluid (CSF) that bathes the brain (see Chapter 2, Box 1) and drains into large venous blood vessels and meningeal lymphatic vessels on the surface of the brain.
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
4 Disease Mechanism/Cause/Basic Science
FIGURE 12 Dual role of γ-secretase cleaving both Notch and
APP. The same enzyme that produces toxic forms of amyloid serves to produce Notch, an important protein that acts as a nuclear gene regulator required for normal development and cell polarity. Notch signaling depends on three endoproteolytic cleavages (S1–S3). Notch maturates in the Golgi by furin-mediated cleavage at site 1 (S1). At the cell surface, Notch is cleaved at S2 (after binding to its ligands Delta/ Serrate/Lag-2). Finally, cleavage at S3 liberates the notch intracellular domain (NICD), which translocates to the nucleus, thereby regulating the transcription of target genes by binding to transcription factors. β-APP is processed by a similar pathway. Initial cleavages of β-APP by α- or β-secretase lead to the generation of a membrane-bound complex. γ-Secretase cleavage then liberates Aβ and the APP intracellular domain (AICD). The biological function of AICD remains to be determined. Generated from Ref. 11.
95
Cell-to-Cell Transmission of Disease The gradual spread of plaques from the mesiotemporal lobe and entorhinal cortex to surrounding cortical regions that ultimately involves essentially all gray matter, has been a puzzle. Recent experiments suggest that Aβ toxicity and plaque formation may spread from cell to cell in a prion-like fashion. Prions are misfolded, self- propagating proteins that have been identified in Scrapie and Creutzfeldt–Jakob disease. A prion entering a healthy cell acts as a template to cause normally folded proteins to misfold in the same way as the template. If Aβ behaved as a prion, it could indeed self-propagate in the brain, spreading from one region to the next. In transgenic mice that overexpress a mutant form of APP, the intracranial injection of Aβ from brain homogenates into only one brain hemisphere indeed caused widespread Aβ deposits in both hemispheres, consistent with a prion-like mode of propagation.13 Since prions are infectious particles that can spread disease from one infected individual to another, a prion-like spread of amyloid could, in theory, make AD a communicable disease. Fortunately, we typically do not come in contact with brains of other people, but such transmission could occur through surgical instruments or corneal transplants.
4.3 Vascular Changes in Alzheimer Disease Next to age and family history, any health condition with vascular disease presents a significant risk factor for AD. This includes diabetes, hypertension, obesity, and stroke. It has been well documented that AD presents with vascular pathology, particularly affecting small vessels in the cortex, causing general reduction in cerebral blood flow or hypoperfusion.14 This alone could negatively affect not only the delivery of glucose and oxygen as energy substrates but also the clearance of Aβ from the cerebrospinal fluid. Lipoprotein receptor-related protein 1 (LRP1) is normally expressed on the abluminal side of cerebral blood vessels, and serves to bind Aβ for its clearance from the brain into the blood across an intact BBB. However, the extensive amyloid deposits along cerebral vessels (Figure 13) compromise vessel integrity, causing a breach in the BBB. This, in turn, will compromise Aβ clearance and indeed may cause further Aβ influx from peripheral blood, along with other harmful reagents, including albumin and glutamate. These molecules may contribute to seizures and excitotoxicity.
4.4 The Genetics of Alzheimer Disease FIGURE 13 Blood vessel association of Aβ visualized in a hAPP mutant mouse in vivo using a fluorescent dye (benzothiazole) that binds to amyloid plaques (white). Courtesy of Ian Kimbrough, School of Neuroscience, Virginia Tech.
The two greatest risk factors for AD are age and a family history of dementia. The heritance of AD appears to differ absolutely between early-onset and late-onset disease. Early-onset AD accounts for 5% of all AD cases, and 1% of cases are caused by rare mutations in APP,
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
96
4. AGING, DEMENTIA, AND ALZHEIMER DISEASE
PSEN1, or PSEN2.2 These mutations essentially guarantee disease onset before age 60. Mutations are largely autosomal dominant and follow Mendelian inheritance. By contrast, late-onset or sporadic AD has no consistent mode of transmission, occurs after the age of 60, and accounts for the majority of all AD cases. Nevertheless, genetic predispositions must play a role in sporadic AD as well, since the disease risk is significantly elevated if first-degree relatives suffer from dementia. The gene variant most strongly linked to an increase in disease risk for late-onset AD is the ε4 allele of the APOE gene. APOE is a lipid-binding lipoprotein that is synthesized in the liver, as well as by astrocytes, microglial cells, and neurons in the brain. APOE binds and transports lipids, including cholesterol. Neurons express the low-density lipoprotein receptors that bind APOE. The three isoforms of APOE differ by only a few amino acids, but these are sufficient to alter their relative affinity for cell surface receptors, with ε2 binding most poorly and ε4 most well. ε4 is expressed in 14% of the population and increases disease risk between 4- and 15-folds. An individual carrying both alleles has a higher risk, whereas an individual with only one ε4 allele present and a family history of AD has a lower risk. By contrast, the ε2 allele, present in 7% of people, confers protection, while the ε3 allele, expressed in 79% of the population, is neutral. Presence of the ε4 allele is correlated with increased amyloid deposition, and it appears to promote the “seeding” of amyloid plaque deposits around blood vessels and reduces the clearance of Aβ at the blood–brain barrier. A number of large genome-wide association studies have searched for additional genes that increase disease risk, and these have identified ∼ 15 other candidates. However, none of these candidates changes risk by more than 10%, minute compared to the 400–1500% of APOE4.
4.5 Animal Models of Alzheimer Disease To study disease pathology and develop drugs for treatment, animal models that recapitulate salient aspects of the disease are essential. The vast majority of AD cases are sporadic and do not have a known cause. Only a small number of patients carry heritable mutation in genes linked to the accumulation of Aβ. The introduction of these genetic mutations has allowed the generation of mice that recapitulate some, yet not all, aspects of the disease. These have been essential to provide support for the amyloid cascade hypothesis. Mice overproducing mutant APP develop extracellular plaques like those found in human autopsies. Increased Aβ42 expression worsens the disease, while increased Aβ40 reduces disease severity. AD mice develop cognitive impairments; however, unlike in humans, these precede rather than trail amyloid deposits. This was the foundation for considering soluble oligomeric amyloid as causative agent
in cognitive decline. A disease element not well replicated is the development of NFTs. As discussed above, tangles are intraneuronal aggregates of hyperphosphorylated tau and are a hallmark of human AD. However, none of the APP-overexpressing mice develop tau pathology. Finally, the inflammatory response observed in human disease differs markedly from that observed in mouse models. Several reasons have been given for these discrepancies between mouse and human AD, not least important of which is the very short life span of a mouse compared to that of humans. In spite of these shortcomings, these mouse models have provided detailed insight into many aspects of disease, such as the role of γ-secretases in Aβ42 production.15 Also, the cellto-cell propagation of disease, whereby Aβ acts in a prion-like fashion, could not have been disclosed without a suitable mouse model. Unfortunately, none of the drug targets that showed promise in mouse models held up in clinical trials. Hence there remains a pressing need to develop more refined disease models that more accurately mimic the salient elements of disease.
4.6 Changes in Transmitter Systems in Alzheimer Disease Many of the cortical neurons affected in AD, and particularly those involved in learning and memory, are glutamatergic. However, cholinergic neurons are important modulators of attention and memory formation. Interestingly, choline acetyltransferase, the enzyme responsible for the neuronal synthesis of acetylcholine (ACh), is reduced by 50–90% in the cortex and hippocampus of AD patients, and the enzymatic loss is roughly proportional to the severity of cognitive loss. Autopsies from AD patients have shown a loss of neurons in the basal nucleus of Meynert, which extends cholinergic projections to the cerebral cortex, particularly frontal lobe areas tasked with executive functions, many of which are compromised in AD. In light of these findings, drugs such as Donepezil or Rivastigmine that increase the levels of acetylcholine by reducing its breakdown via cholinesterases are frequently used to treat early, modest deficits in AD. Furthermore, clinical trials discussed at the end of this chapter suggest that protection of cholinergic neurons from cell death hold promise in slowing memory decline in patients. Finally, all of the major modulators of mood, anxiety, and behavior (namely, noradrenaline, serotonin, and dopamine) are also significantly reduced in tissue from AD patients, explaining some of the changes in mood and behavior that characterize late stages of AD.
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
97
5 Treatment/Standard of Care/Clinical Management
4.7 Alzheimer Disease and Inflammation Inflammation is common among acute and chronic nervous system diseases, and AD is no exception. Neuroinflammation encompasses the activation of microglia, the resident immune cells of the brain that derive from macrophages, as well as activation of astrocytes. Each can release proinflammatory molecules and cytokines such as IL-1β, IL-6, and TNFα, as well as complement components. Amyloid deposits are sufficient to activate both microglia and astrocytes in culture and both appear to engage in the removal of Aβ. It has been suggested that early in disease, inflammation may be beneficial and facilitate Aβ clearance, whereas late in disease, neuroinflammation may be detrimental. Here, these same molecules begin to interfere with neuronal function, negatively affecting synaptic transmission and the stability of cell processes. The production of reactive oxygen species (ROS) by microglial cells is sufficient to cause process retraction, and chronic increases in ROS can induce apoptosis. It is possible that a significant aspect of the progressive disease pathology is mediated by reactive astrocytes and activated microglial cells, which, rather than supporting neuronal function, turn into deadly vices. Unfortunately, the immune responses seen in AD patients are poorly replicated in animal models of disease and hence difficult to study. One observation that confirms a role of inflammation in AD is the positive effect of nonsteroidal antiinflammatory drugs (NSAIDs) such as ibuprofen, which reduce Aβ burden and, in humans, reduce the likelihood of disease. The beneficial effect appears to be limited to those NSAIDs that inhibit γ-secretase while also attenuating neuroinflammation.16 Note, however, that metaanalysis of currently available clinical trials, in spite of some re-
ported successes, does not support the use of currently available NSAIDs to treat AD.17
5 TREATMENT/STANDARD OF CARE/ CLINICAL MANAGEMENT Most dementias, including AD, are progressive and untreatable, leaving the clinician primarily to manage disease symptoms. However, some dementias that result from vitamin deficiencies or drug exposure can be effectively treated and even cured, and hence it is essential to reach the most accurate diagnosis for any given patient to ensure the most adequate support of patient and caregiver. Patient history and behavioral presentation are often of highest diagnostic value. For example, an elderly patient who has experienced a slow and progressive decline in episodic memory function is most likely to suffer from AD. By comparison, a younger person who has become compulsive, suffering poor judgment and disinhibition, yet without apparent memory loss, is more likely to suffer from FTD. Sudden onset and a history of stroke would be highly suggestive of vascular dementia. The neurological evaluation will also search for any evidence of motor involvement, which would be uncommon in AD, yet a frequent feature in FTD and DLB. The Mini-Mental Status Examination (Table 1) is a convenient way to assess early memory deficits. Difficulties recalling names or numbers from a list are early signs of AD, and yet are not affected by FTD. These and other distinguishing features for the most common forms of dementia are summarized in Table 2.
TABLE 2 Distinguishing features for the most common forms of dementia. Disease
Initial symptoms
Cognitive and psychiatric symptoms
Motor symptom
Imaging hallmarks
AD
Loss of memory
Loss of episodic memories
Initially unimpaired
Atrophy of hippocampus and entorhinal cortex
FTD
Apathy; impaired judgment and/or speech; hyperorality
Frontal and executive impairments, euphoria, depression
Palsy, rigidity, dystonia
Atrophy of frontal, insular, and temporal lobes, but spares posterior parietal lobe
DLB
Hallucinations and/or Capgras’ syndrome, sleep impairments
Frontal/executive impairments, prone to delirium, but no impairment to memory
Parkinsonism
Atrophy of posterior parietal atrophy, lacks severity of hippocampal atrophy seen in AD
CJD
Dementia, mood/anxiety, movement impairments
Variable frontal and executive impairments, memory, depression, anxiety
Rigidity, Parkinsonism
Ribboning of cortex, hyperintensity of basal ganglia/ thalamus
Vascular
Typically sudden onset, but precise symptoms vary; falls, weakness
Slowing of frontal/executive and cognitive function, delusions, anxiety, but may spare memory
Slowing of movement, spasticity
Infarctions of cortical or subcortical areas, white matter changes
AD, Alzheimer Disease; FTD, frontotemporal dementia; DLB, dementia with lewy bodies; CJD, Creuztfeldt Jacobs Disease; Vascular, vascular dementia.
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
98
4. AGING, DEMENTIA, AND ALZHEIMER DISEASE
5.1 Imaging, Diagnosis, and Disease Prediction Structural imaging plays a major role in the diagnosis of dementias and is now considered standard of care.18 Firstly, MRI or even computed tomography imaging is useful in establishing whether dementia may be the consequence of a brain insult, such as an infarct, tumor, or infection. Secondly, different imaging modalities can provide complementary information regarding diagnosis, differential diagnosis, and monitoring of disease progression. Most readily available are standard, structural MRI scans, which permit assessment of global or focal atrophy. Examples comparing two 75-year-old subjects, one with AD and the other without, show a striking atrophy of the hippocampus and surrounding mesiotemporal lobe in the AD patient (Figure 14). Longitudinal imaging studies suggest that even mildly affected individuals have a 20–30% decrease in entorhinal and 15–20% decrease in hippocampal volume. The rate of decrease is estimated to be 3–5% per year, and a decrease in hippocampal volume of approximately 10% can already be detected 3 years prior to disease symptoms. The major advantages of structural MRI over other imaging techniques are availability, cost, reproducibility, and quantitative nature of the readout. A second imaging modality that shows brain activity more directly is FDG-PET. In this modality, fluordeoxyglucose (FDG), a glucose analogue, can be detected by positron emission tomography (PET) when labeled with radioactive isotopes such as fluorine-18. Since the brain’s
energy use depends almost exclusively on glucose, the consumption of this tracer reports the resting metabolic activity of the brain, primarily attributable to synaptic activity. The drawbacks of FDG-PET are cost and availability, and it is sensitive to potential erroneous contribution of other metabolic changes. However, it is also a powerful tool to distinguish AD from FTD. As shown in Figure 15, reduced glucose consumption in the temporal and occipital lobe characterizes AD, whereas selective loss in the frontal lobe signifies FTD. The most recent addition to the imaging arsenal, often considered a “game changer,” is amyloid PET. Here, the patient receives a PET tracer that specifically binds to the beta sheet structure of fibrillary, insoluble Aβ. The most commonly used compound is called PiB or “Pittsburg compound.” It is a radioactive derivative of thioflavin T that has long been used to label beta sheet structures such as amyloid in histological studies. Of note, these tracers do not detect oligomeric Aβ. Their predictive value for amyloid deposits or cerebral amyloidosis is very high. Fifteen studies that independently used this approach found positive amyloid signal in 96% of all patients (Figure 16). Moreover, when applied to patients with mild cognitive decline, who were followed up longitudinally for 3 years, 57 of 155 patients progressed to AD, and an impressive 53 of the 57 were amyloid positive. Only 7% of the a myloid-negative patients developed AD.18 Unfortunately, amyloid PET is even less widely available than FDG-PET and is very expensive. This may change with the use of labels with a lon-
FIGURE 14 Structural changes detected in Alzheimer Disease by noninvasive imaging. Examples of structural T1-weighted MRI comparing two 75-year-old individuals, one control (left) to one with AD (right) showing marked mesiotemporal lobe atrophy (circle). Reproduced with permission from Ref. 6.
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
5 Treatment/Standard of Care/Clinical Management
99
FIGURE 15 FDG-PET used to noninvasively distinguish patients suffering from AD and FTLD. In AD, glucose utilization determined by FDG-PET is lowest in the occipital and parietal lobes of the cortex (arrows), while in FTD, the frontal lobe shows the greatest reduction in glucose utilization. Courtesy of Drs Frederik Barkhof, Marieke Hazewinkel, Maja Binnewijzend, and Robin Smithuis, Alzheimer Center and Image Analysis Center, Vrije University Medical Center, Amsterdam and the Rijnland Hospital, Leiderdorp, The Netherlands. http://www.radiologyassistant.nl/data/bin/ a509797720938b_FDG-pet.jpg.
ger half-life, such as fluorine-18, which does not require an on-site cyclotron for production.
5.2 Drug Treatment We currently lack drugs that can prevent or cure dementia, so treatments target symptoms and attempt to reduce the burden on the patient and caregiver. Commonly used drugs include acetylcholinesterase inhibitors such as tacrine, donepezil, rivastigmine, and galantamine, which prolong the activity of ACh at the synaptic cleft, thereby mitigating the overall loss of ACh in cholinergic synapses in AD. This class of drugs has shown small benefits in the cognitive per-
formance of many patients, and, given their low cost, any one of them is recommended for management of mild-moderate AD. A second approved drug is memantine, a use-dependent blocker of NMDA receptors. In clinical studies, it has shown limited benefits to a select group of patients. Since depression is a frequent comorbidity of dementia, antidepressants, particularly selective serotonin reuptake inhibitors such as escitalopram (Lexapro), are useful. Similarly, psychotropic drugs such as quetiapine (Seroquel) can help reduce delusions and psychosis in late stages of dementia.
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
100
4. Aging, Dementia, and Alzheimer Disease*
FIGURE 16 Patient diagnosis using the Pittsburg compound which directly binds to amyloid deposits in the brain of AD patients. The example illustrates abundant amyloid labeling in the AD patient not seen in the control. Courtesy of Prof. Rowe and Villemagne, Austin Health, Australia.
5.3 Disease Onset Delay and Prevention In the absence of disease-modifying therapies, it behooves us to consider disease prevention as a means to control the burden of disease to the individual and to society. Delaying disease onset by just a few years, even for a small number of people, would have a profoundly positive influence and may enable many to reach the end of their normal life without succumbing to dementia. Delaying the onset of dementia by 5 years would reduce worldwide dementia cases by 50%.19 How could this be achieved? Obviously, the greatest risk factor for dementia is age, which is a nonmodifiable risk factor. The same is true for ApoE4 status. However, epidemiological studies suggest that up to 35% of all AD cases are attributable to modifiable risk factors20 and hence may be preventable.19 The leading modifiable risk factors are educa-
tional status, physical inactivity, smoking, social isolation and metabolic issues including obesity and diabetes. Interestingly, risk factor profiles differ in the United States compared to the rest of the world. In the United States, the leading risk factors for AD are physical inactivity followed by depression and smoking, whereas globally, low education leads over smoking and physical inactivity. As these risk factors are additive, smoking cessation combined with increased physical activity alone could delay individual symptom onset and have profound effects on the total number of individuals living with AD. Preventative measures include exercise and physical activity, antihypertensive drugs, cognitive interventions, social engagement, dietary changes, and education. Some of these are easy to implement while others are more difficult. Importantly, interventional strategies
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
5 Treatment/Standard of Care/Clinical Management
101
BOX 2
THE NUN STUDY Sister Matthia lived to be 104 years old and was laughing and knitting socks just weeks before her death (Box 2-Figure 1). Her fellow sister Agnes, by contrast, developed dementia in her 80s and became so cognitively impaired that she lost all sense of language and only communicated through an occasional smile. What both had in common was much more than what differentiated them. Both joined the Sisters of Notre Dame convent in Minnesota in their early 20s. Both taught elementary school students throughout life. They had similar rooms in almost identical buildings, got up at the same time, ate the same food, and, by and large, spent the majority of their time doing similar things. So did 678 of their fellow sisters who were part of the “Nun” study, a longitudinal study examining risk factors for dementia and Alzheimer Disease. All nuns agreed to annual physical and cognitive exams. They shared their medical data and ultimately donated their brains to science. The sisters devoted their lives to God and serving the community. They never married or had children, and many environmental and lifestyle factors including alcohol, smoking, and drug use were largely absent from their lives. Hence the sisters represented a very homogeneous population. So, what did scientist learn? Despite living in similar environments for the majority of their lives, the Sisters differed widely in their cognitive ability and overall physical health. Their health ranged from some 75-year-old nuns being so severely demented that they could barely utter a word, to 100-year-old women who were sharp as a tack and in outstanding physical health. The histological analysis of the sisters’ brains also varied but surprisingly showed that the amyloid and tau burden in the brain often did not correlate with dementia as strongly as had been assumed previously. For example, Sister Marcella scored a 28/30 on her MME at age 100 and had no visible brain abnormalities, yet Sister Matthia who became the “gold standard” for health aging in perfect cognitive health at age 104 yet had abundant AD pathology scoring a 4 on the Braak scale. Sister Bernadette who died of a heart attack at age 85 had nearly perfect mental scores when examined on her 81st through 85th birthdays, yet her brain had the highest pathological burden with a Braak stage of 6. She also carried two of the APOe4 alleles, placing her at high risk of developing AD. So, what distinguished those that developed AD from those who did not? When scientists examined the autobiographical essays submitted by the nuns as they entered the convent at age
18–25, they found that those applicants that had a greater verbal fluency or idea density, measures for language ability or sophisticated word use, had a reduced risk of developing AD while those lacking in language ability were at highest risk of developing AD later in life. Here are two examples that illustrate this difference in verbal fluency: Sister Nicolette, who did not develop AD wrote: “After I finished the eighth grade in 1921 I desired to become an aspirant at Mankato but I myself did not have the courage to ask the permission of my parents so Sister Agreda did it in my stead and they readily gave their consent.” On the same topic, Sister Mathilda, who had profound AD had written: “After I left school, I worked in the post office.” Scientists reason that an explanation for the correlation of language and AD may be a greater “mental reserve” in individuals who have better education, verbal skills and such, allowing them to resist a greater degree of brain pathology. Alternatively, language development may be linked in still unknown ways to amyloid and tau biology. Regardless, your Middle and High School teachers, who are at least partially responsible for your linguistic ability, may be more important for a future healthy life than you may have thought!
BOX 2-FIGURE 1 Sister Matthia Gores. Courtesy of the School Sisters of Notre Dame North American Archives, Milwaukee, Wisconsin.
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
102
4. Aging, Dementia, and Alzheimer Disease*
must occur long before disease onset. For example, less education is associated with a 60% increase in relative dementia risk. It is generally attributed to higher cognitive reserve, which refers to the ability of an individual to afford a greater loss of neuronal function and still be relatively cognitively intact. Such intervention would have to occur in school-aged children. Note that the concept of cognitive reserve may also explain findings of the Nun study (Box 2) where verbal fluency in early adulthood was most predictive of future AD. Hypertension and metabolic intervention are likely of greatest importance to treat in midlife (45–65 years of age) whereas physical activity and diet may be beneficial throughout life.
6 EXPERIMENTAL APPROACHES/ CLINICAL TRIALS Given the gravity of this disease, and the tremendous number of patients affected worldwide, it is not surprising that there are numerous efforts under way to explore new treatment strategies. There are currently
over 430 active clinical trials in the United States that include interventional studies, studies on early d etection, or the use of symptomatic treatment of comorbidity. Of the 144 interventional studies, 44 are early safety (phase 1) trials; 79 are early efficacy (phase 2) trials; and 48 are later stage (phase 3–4) trials. Some of the phase 1 and 2 are double counted as they share both safety and efficacy as outcome. Only a very limited number of trials actually study truly novel molecules for treatment, and we will highlight a few examples below. Given what we just learned about the etiology of disease, any strategy to reduce Aβ or tau burden would appear to be the most likely to succeed. Indeed, the broad consensus that toxic Aβ deposits are central to the development of AD has led to several approaches aimed at reducing the Aβ burden in AD patients, with the various angles taken toward this problem schematically illustrated in Figure 17. A first attempt involved a vaccine that detects and destroys Aβ. This clinical trial came on the heels of very promising studies in animal models and positive safety data in a phase 1 human study. However, a phase 2 vaccine trial using a synthetic Aβ42 (AN1792,
FIGURE 17 Future treatments for Alzheimer Disease are exploring four major targets. (1) The cleavage of APP into plaque-prone amyloid by inhibition of the β- or (2) γ-secretases (β and γ symbol). (3) Interference of plaque formation using inhibitors of Ab. (4) Enhancing the clearance of Ab using immunotherapy. Reproduced with permission from Ref. 10.
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
6 Experimental Approaches/Clinical Trials
Eli Lilly) had to be terminated because five enrolled patients suffered severe brain inflammation.21 Follow-up examination of brain tissue revealed significant microglial activation, reactive gliosis, and amyloid angiopathy, all consistent with widespread inflammation. Instead of inducing the body to produce antibodies through a vaccine, the studies that followed used humanized monoclonal antibodies, such as bapineuzumab and solanezumab, that directly bind Aβ42 instead. Although these were well tolerated by patients, the results from large phase 3 clinical trials were disappointing for both drugs,22 as neither demonstrated any improvement of cognitive outcome or daily living. Rather than removing already synthesized Aβ42 from the brain and blood, an alternative approach is inhibiting additional Aβ42 production. As this requires sequential cleavage of APP by β- and γ-secretases, a number of inhibitors have been developed targeting either of the two enzymes. Three beta β-amyloid precursor protein-cleaving enzyme (BACE) inhibitors (verubecestat, elenbecestat, and aducanumab) were examined in phase III clinical trials by different companies and together enrolled well over 3000 patients with mild cognitive impairment or early AD. Unfortunately, none of the trials showed efficacy in slowing disease progression and all three were terminated early. Trials with γ-secretase inhibitors fared even worse. A phase III study with semagacestat was terminated early because patient symptoms worsened while on the drug. As alluded to before, one complication with γ-secretases is their role in cleaving other membrane-embedded proteins such as the Notch receptor, which acts as an important signaling molecule. In the skin, Notch acts as a tumor suppressor gene, and blockade of γ-secretase would create a cancer risk. Indeed the γ-secretase inhibitor LY450139 (Eli Lilly) and Avagacestat (Bristol-Myers Squibb) failed
103
for this very reason. Since then, new drugs have been synthesized that, rather than inhibiting γ-secretase activity, modify its action such that the cleavage produces the more soluble Aβ38 instead of the toxic insoluble Aβ42. As mentioned before, some NSAIDs, for example, ibuprofen, change the γ-secretase cleavage of Aβ, and d erivatives of these γ-secretase modulators are now in various stages of clinical testing as well. Interestingly, long-term chronic users of ibuprofen are 44% less likely to develop AD,23 which is consistent with the hypothesized reduction in toxic Aβ42. However, ibuprofen does not ameliorate disease in patients already diagnosed with AD. New drugs are in development that inhibit APP cleavage selectively while largely sparing Notch. The best of these have a several 100-fold higher activity on Notch cleavage versus APP. How effective these γ-secretase modulators are will be understood at the conclusion of ongoing clinical trials.24 The main reason that the above trials lacked effectiveness may have been timing. The slowing of aggregates or clearance of Aβ occurs much too late in the disease, when disease is already rampant and much of the neuronal and cognitive damage is already done. The research we discussed above suggests, as schematically illustrated in Figure 18, that the pathogenic cascade of AD begins at least one or even two decades before cognitive impairments manifest, during which accumulations of Aβ42 and phosphorylated tau cause progressive brain atrophy. If one could introduce therapies to reduce the Aβ42 burden prior to disease onset, one may be able to delay or even prevent dementia. Hence a necessary first step toward an early disease intervention is the reliable identification of persons at risk of developing AD. As we discussed above, this is now within reach using a combination of biomarkers. The currently established markers include structural MRI, which examines regions of global brain shrinkage;
FIGURE 18 Schematic depiction of disease progression. Ab aggregation into plaques precedes disease symptoms by as many as 15 years, while tau hyperphosphorylation may occur up to 5 years prior to symptoms. As pathology develops, mild cognitive impairment (MCI) eventually gives way to AD-dementia, a stage at which pathological signs of tissue changes are abundant.
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
104
4. AGING, DEMENTIA, AND ALZHEIMER DISEASE
FDG-PET to measure a decline in glucose consumption; measurement of fibrillary Aβ using an amyloid PET ligand; and cerebrospinal fluid (CSF) or plasma levels of Aβ42 and phosphorylated tau. Although none of these is 100% predictive, in combination, they provide a high level of confidence.25 Clinical studies that evaluated the ability to detect AD risk early using these biomarkers have focused on two groups: those carrying two APOE ε4 alleles and that hence have up to 15-fold increased AD risk, and those with disease-causing mutations in either PSEN1/2 or APP, who have a near 100% likelihood to develop the disease. The latter group in particular provides a powerful sample as, for each individual, given sufficient time, AD symptoms will develop for sure. Studies to date suggest that mutation carriers present with a reduction in hippocampal volume ∼ 15 years before symptom onset, changes in glucose utilization precede symptoms by ∼ 10 years, and PET studies detecting fibrillary Aβ similarly show deposits ∼ 10–25 years prior to clinical disease onset. Tau pathology, judged from changes in CSF samples, also significantly precedes disease. The picture that emerges (Figure 18) is a sequential acquisition of pathology during a presymptomatic stage of disease. Taking advantage of early detection, current efforts are exploring disease intervention strategies where atrisk individuals, biologically defined, are enrolled in trials for the very Aβ monoclonal antibodies that had failed in patients with already established dementia. A large proof-of-concept study has just been launched through a corporate– government joint venture. The National Institutes of Health is participating in a $100 million disease prevention trial in which 200 subjects belonging to a Colombian kindred with the heritable PSEN1 E280A mutation will be enrolled. This mutation reliably causes cognitive impairments by age 44. The trial (www.clinicaltrials.gov, identifier NCT01998841) will evaluate cognitive status in 100 subjects treated with the antibody crenezumab (Genentech) prior to disease onset and throughout its course. These patients will be compared to 100 subjects receiving a placebo. If successful, this approach could be expanded to include other at-risk patients, for example, those presenting with APOE4 with evidence of amyloid deposits detected by amyloid PET imaging. Unless any of these disease prevention trials succeeds, the ability to detect disease early leaves us with an ethical dilemma. If applied to the general population, at significant cost to our healthcare system, a combination of biomarkers will allow us to identify persons at high risk to develop AD a decade or two prior to disease onset. What will we do with this knowledge? If we had treatments to offer that could delay or even prevent d isease, such tests would be well justified. However, in the absence of therapy, these
evaluations may become entirely academic and indeed may even be psychologically harmful to the patient. This should be taken seriously by companies offering mail-order genetic testing. A different approach that does not target amyloid, but instead tries to retain cognitive function by boosting the survival of cholinergic neurons, is beginning to show promise as well.25, 26 It has long been recognized that over the course of AD, patients lose up to 90% of their cholinergic neurons in the basal forebrain. These neurons project to all areas of the cerebral cortex and hippocampus and modulate synaptic activity involved in learning and memory. Nerve growth factor (NGF) prevents cholinergic neuronal death and enhances memory function in mouse models of AD. Unfortunately, NGF is difficult to deliver, as it does not pass the BBB, and intracranial injection causes severe side effects by acting on neurons not involved in the memory circuits. An elegant w ork-around was to harvest skin fibroblasts from AD patients and genetically modify them to produce and release NGF. These were then stereotaxically implanted into the nucleus basalis, the main site of cholinergic neurons, where these cells took hold and began to release NGF. In a small phase 1 clinical trial involving eight patients, this approach was not only safe but also produced marked and long-lasting improvements in cognition and significantly enhanced metabolic activity judged by FDG-PET imaging (Figure 19). The ultimate utility of this exciting approach will have to await further clinical testing in larger patient cohorts.
7 CHALLENGES AND OPPORTUNITIES Dementia may be among the most frightening conditions a human can experience. Memories are the substrate of learning and cognition, the very fabric of humanity. Improved health care, nutrition, and lifestyle changes now allow us to live decades longer than our ancestors, and our offspring will soon push life expectancy beyond 100 years. Yet, if we cannot find a cure for dementia, this may not be desirable at all. Given the near-exponential rise in disease incidence with age, most individuals would be guaranteed to suffer from AD in their last decade(s) of life. Some argue that dementia is inevitably linked to aging. If true, we must find ways to slow down aging itself rather than be concerned about individual molecular processes of age-related disease. In the near term, however, dementia is a health epidemic and is bound to break our healthcare system if success does not come soon. The cost to society and the emotional burden on caregivers will be unsustainable. Fortunately, a large number of researchers have devoted their lives to the study of dementia. As illustrated throughout this chapter, our
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
7 Challenges and Opportunities
105
FIGURE 19 Averaged FDG PET scans in four subjects treated with NGF, overlaid on standardized MRI templates. Representative axial sections, with 6–8 months between first and second scan, showing widespread interval increases in brain metabolism. Flame scale indicates FDG use/100 g tissue/min; red color indicates more FDG use than blue. Reproduced with permission from Ref. 26.
understanding has improved in leaps and bounds and has done so within a relatively short amount of time. However, significant gaps remain in our understanding. For example, we still do not know whether it is the amyloid deposits that are toxic to neurons, or whether instead oligomeric amyloid is the culprit. Similarly, whether tau inclusions are harmful or protective is unclear. Does the prion hypothesis for disease spread hold up to future testing? What is the contribution of
cell-autonomous versus noncell-autonomous processes to disease? Is there a role of astrocytes, microglial cells, and complement components? How about the immune system’s role in disease? Breaches in the BBB may significantly c ontribute to disease, allowing entry of harmful serum components, lymphocytes, and macrophages. The peculiar time course of disease suggests that it begins at least 10, if not 20, years prior to the first signs of dementia. Does that represent the time window of
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
106
4. AGING, DEMENTIA, AND ALZHEIMER DISEASE
opportunity to interfere? If so, are there modifiable risk factors that one could address to mitigate or delay disease onset? Neurogenetics is still in its infancy, and the ability to sequence whole genomes and search through genomic data from thousands of patients is on the horizon. Will such large genomic screening identify the still missing genes to explain sporadic late-onset AD? One can only hope that answers will come soon!
Acknowledgments This chapter was graciously reviewed by Erik Roberson, MD, PhD, the Rebecca Gale Endowed Professor in the Departments of Neurology and Neurobiology. He is also the Director of Alzheimer’s Disease Center as well as the Center for Neurodegeneration and Experimental Therapeutics at the University of Alabama at Birmingham.
References 1. Glenner GG, Wong CW. Alzheimer’s disease: initial report of the purification and characterization of a novel cerebrovascular amyloid protein. Biochem Biophys Res Commun. 1984;120:885–890. 2. Tanzi RE. The genetics of Alzheimer disease. Cold Spring Harb Perspect Med. 2012;2. https://doi.org/10.1101/cshperspect. a006296. 3. Seshadri S, et al. The lifetime risk of stroke: estimates from the Framingham Study. Stroke. 2006;37:345–350. https://doi. org/10.1161/01.STR.0000199613.38911.b2. 4. Lourida I, et al. Mediterranean diet, cognitive function, and dementia: a systematic review. Epidimiology. 2013;24:479–489. https://doi.org/10.1097/EDE.0b013e3182944410. 5. Landau SM, et al. Association of lifetime cognitive engagement and low beta-amyloid deposition. Arch Neurol. 2012;69:623–629. https://doi.org/10.1001/archneurol.2011.2748. 6. Scheltens P. Imaging in Alzheimer’s disease. Dialogues Clin Neurosci. 2009;11:191–199. 7. Iqbal K, Liu F, Gong CX, Grundke-Iqbal I. Tau in Alzheimer disease and related tauopathies. Curr Alzheimer Res. 2010;7:656–664. 8. Hartmann T, et al. Distinct sites of intracellular production for Alzheimer’s disease A beta40/42 amyloid peptides. Nat Med. 1997;3:1016–1020. 9. Mucke L, Selkoe DJ. Neurotoxicity of amyloid beta-protein: synaptic and network dysfunction. Cold Spring Harb Perspect Med. 2012;2:a006338. https://doi.org/10.1101/cshperspect.a006338. 10. Roberson ED, Mucke L. 100 Years and counting: prospects for defeating Alzheimer’s disease. Science. 2006;314:781–784. https:// doi.org/10.1126/science.1132813. 11. Walter J, Kaether C, Steiner H, Haass C. The cell biology of Alzheimer’s disease: uncovering the secrets of secretases. Curr Opin Neurobiol. 2001;11:585–590. https://doi.org/10.1016/ S0959-4388(00)00253-1. 12. Mattson MP. Neurobiology: Ballads of a protein quartet. Nature. 2003;422:385. 387 https://doi.org/10.1038/422385a. 13. Stohr J, et al. Purified and synthetic Alzheimer’s amyloid beta (Abeta) prions. Proc Natl Acad Sci U S A. 2012;109:11025–11030. https://doi.org/10.1073/pnas.1206555109. 14. Sagare AP, Bell RD, Zlokovic BV. Neurovascular dysfunction and faulty amyloid β-peptide clearance in Alzheimer disease. Cold Spring Harb Perspect Med. 2012;2. https://doi.org/10.1101/cshperspect.a011452. 15. LaFerla FM, Green KN. Animal models of Alzheimer disease. Cold Spring Harb Perspect Med. 2012;2. https://doi.org/10.1101/cshperspect.a006320.
16. Saura CA. Presenilin/gamma-secretase and inflammation. Front Aging Neurosci. 2010;2:16. https://doi.org/10.3389/ fnagi.2010.00016. 17. Jaturapatporn D, Isaac MG, McCleery J, Tabet N. Aspirin, steroidal and non-steroidal anti-inflammatory drugs for the treatment of Alzheimer’s disease. Cochrane Database Syst Rev. 2012;2:Cd006378. https://doi.org/10.1002/14651858.CD006378.pub2. 18. Johnson KA, Fox NC, Sperling RA, Klunk WE. Brain imaging in Alzheimer disease. Cold Spring Harb Perspect Med. 2012;2. https:// doi.org/10.1101/cshperspect.a006213. 19. Livingston G, et al. Dementia prevention, intervention, and care. Lancet. 2017;390:2673–2734. https://doi.org/10.1016/ s0140-6736(17)31363-6. 20. Barnes DE, Yaffe K. The projected effect of risk factor reduction on Alzheimer’s disease prevalence. Lancet Neurol. 2011;10:819–828. https://doi.org/10.1016/s1474-4422(11)70072-2. 21. Birmingham K, Frantz S. Set back to Alzheimer vaccine studies. Nat Med. 2002;8:199–200. https://doi.org/10.1038/nm0302-199b. 22. Mullard A. Sting of Alzheimer’s failures offset by upcoming prevention trials. Nat Rev Drug Discov. 2012;11:657–660. 23. Vlad SC, Miller DR, Kowall NW, Felson DT. Protective effects of NSAIDs on the development of Alzheimer disease. Neurology. 2008;70:1672–1677. https://doi.org/10.1212/01. wnl.0000311269.57716.63. 24. De Strooper B, Iwatsubo T, Wolfe MS. Presenilins and gamma-secretase: structure, function, and role in Alzheimer disease. Cold Spring Harb Perspect Med. 2012;2. https://doi. org/10.1101/cshperspect.a006304, a006304. 25. Tuszynski MH, et al. Nerve growth factor gene therapy: activation of neuronal responses in Alzheimer disease. JAMA Neurol. 2015;72:1139–1147. https://doi.org/10.1001/ jamaneurol.2015.1807. 26. Tuszynski MH, et al. A phase 1 clinical trial of nerve growth factor gene therapy for Alzheimer disease. Nat Med. 2005;11:551–555. https://doi.org/10.1038/nm1239.
General Readings Used as Source 1. Global, regional, and national burden of Alzheimer’s disease and other dementias, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurol. 2019;18:88–106. https://doi.org/10.1016/s1474-4422(18)30403-4. 2. Livingston G, et al. Dementia prevention, intervention, and care. Lancet. 2017;390:2673–2734. https://doi.org/10.1016/ s0140-6736(17)31363-6. 3. Harrison’s Online, Dementia [chapter 371] by William W. Seeley, Bruce L. Miller (Harrison’s online requires a subscription. Many medical libraries make this available for the students and faculty). 4. Holtzman DM, Mandelkow E, Selkoe DJ. Alzheimer disease in 2020. A collection of articles written by leading experts in the field, and freely available online: http://perspectivesinmedicine.org/ cgi/collection/the_biology_of_Alzheimer_disease. 5. Langbaum JB, Fleisher AS, Chen K, et al. Ushering in the study and treatment of preclinical Alzheimer disease. Nat Rev Neurosci. 2013. http://www.nature.com/nrneurol/journal/v9/n7/full/ nrneurol.2013.107.html. 6. Zerr I. Understanding Alzheimer’s Disease; 2013. This is an open access online multi-author book: http://www.intechopen.com/ books/understanding-alzheimer-s-disease. 7. Alzheimer’s disease, facts and figures. Alzheimers Dement. 2019;15(3):321–387. https://www.alz.org/alzheimers-dementia/ facts-figures. 8. Yamazaki Y, Zhao N, Caulfield TR, Liu CC, Bu G. Apolipoprotein E and Alzheimer disease: pathobiology and targeting strategies. Nat Rev Neurol. 2019;15:501–518. https://doi.org/10.1038/s41582-019-0228-7.
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
REFERENCES
9. David Sweatt J, ed. Mechanisms of Memory. 2nd ed. Academic Press; 2010. 10. Global, regional, and national burden of Alzheimer’s disease and other dementias, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurol. 2019;18:88–106. https://doi.org/10.1016/s1474-4422(18)30403-4.
Suggested Papers or Journal Club Assignments Clinical Paper 1. Bateman RJ, et al. Clinical and biomarker changes in dominantly inherited Alzheimer’s disease. N Engl J Med. 2012;367(9):795–804. 2. Mayeux R. Clinical practice. Early Alzheimer’s disease. N Engl J Med. 2010;362(23):2194–2201 [Case study]. 3. Karch CM, et al. Expression of novel Alzheimer’s disease risk genes in control and Alzheimer’s disease brains. PLoS One. 2012;7(11):e50976. 4. Naj AC, et al. Common variants at MS4A4/MS4A6E, CD2AP, CD33 and EPHA1 are associated with late-onset Alzheimer’s disease. Nat Genet. 2011;43(5):436–441. 5. Tuszynski MH, et al. Nerve growth factor gene therapy: activation of neuronal responses in Alzheimer disease. JAMA Neurol. 2015;72:1139–1147. https://doi.org/10.1001/ jamaneurol.2015.1807.
107
6. Brickman AM, Khan UA, Provenzano FA, et al. Enhancing dentate gyrus function with dietary flavanols improves cognition in older adults. Nat Neurosci. 2014;17(12):1798–1803. 7. Jaunmuktane Z, et al. Evidence for human transmission of amyloid-beta pathology and cerebral amyloid angiopathy. Nature. 2015;525:247–250. https://doi.org/10.1038/nature15369. 8. Sevigny J, et al. The antibody aducanumab reduces Abeta plaques in Alzheimer’s disease. Nature. 2016;537:50–56. https://doi. org/10.1038/nature19323. 9. Egan MF, et al. Randomized trial of verubecestat for prodromal Alzheimer’s disease. N Engl J Med. 2019;380:1408–1420. https:// doi.org/10.1056/NEJMoa1812840. 10. Schindler SE, et al. High-precision plasma beta-amyloid 42/40 predicts current and future brain amyloidosis. Neurology. 2019. https://doi.org/10.1212/wnl.0000000000008081.
Basic Science Paper 1. Verret L, et al. Inhibitory interneuron deficit links altered network activity and cognitive dysfunction in Alzheimer model. Cell. 2012;149(3):708–721. 2. Stohr J, et al. Purified and synthetic Alzheimer’s amyloid beta (Abeta) prions. Proc Natl Acad Sci U S A. 2012;109(27):11025–11030. 3. Griciuc A, et al. Alzheimer’s disease risk gene CD33 inhibits microglial uptake of amyloid beta. Neuron. 2013;78(4):631–643. 4. De Jager PL, Srivastava G, Lunnon K, et al. Alzheimer’s disease: early alterations in brain DNA methylation at ANK1, BIN1, RHBDF2 and other loci. Nat Neurosci. 2014;17(9):1156–1163.
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
This page intentionally left blank
C H A P T E R
5 Parkinson Disease* Harald Sontheimer O U T L I N E 1
Case Story
109
2 History
110
3
4
Clinical Presentation, Diagnosis, and Epidemiology 3.1 Epidemiology and Risk Factors 3.2 Disease Presentation and Diagnosis 3.3 Disease Stages 3.4 Related Movement Disorders
111 111 112 112 114
Disease Mechanism/Cause/Basic Science 115 4.1 Dopamine and the Control of Movement 115 4.2 The Basal Ganglia 116 4.3 Lewy Bodies Pathology and Progression of Disease 117 4.4 How Might α-Synuclein Aggregation in Lewy Bodies Contribute to Disease? 119 4.5 How Might We Explain the Progressive Neuronal Loss Primarily in the SN? 120 4.6 Why Does It Take 60 Years or Longer for PD to Develop? 122 4.7 Genetics of PD 122 4.8 Animal Models of PD 124
1 CASE STORY Anna’s 24th birthday was coming up and she hated even the thought of it. She remembered a time when she loved parties, having friends at her house, and being the center of attention. No longer! That all changed a year earlier, when her hand started shaking. She first noticed
5
Treatment/Standard of Care/Clinical Management 126 5.1 Dopamine and Dopamine Replacement Therapy 126 5.2 Deep Brain Stimulation 127 5.3 Neuroprotection 128 5.4 Treatment of Nonmotor Symptoms 128
6
Experimental Approaches/Clinical Trials 129 6.1 Nicotine 129 6.2 Urate 129 6.3 Nonsteroidal Antiinflammatory Drugs (NSAIDs) 129 6.4 Ca2 + Channel Blockers 129 6.5 Prevention of Lewy Body Formation 130 6.6 Cell-Based Therapies 130
7
Challenges and Opportunities
131
Acknowledgment
131
References
131
General Readings Used as Source
132
Suggested Papers or Journal Club Assignments Clinical Paper Basic Paper
132 132 133
it in Christmas evening when she helped grandmother set the table. Holding just a fork, her right hand trembled. She was worried. Unable to stop the movement, she let go of the fork and pushed her hand forcefully on the table. “What was that?” her mom asked. “I have no idea,” Anna replied. Afraid it would happen again, she only used her left hand to eat. As everyone settled in for the
* The term “Parkinson Disease” is often incorrectly used as possessive noun with an apostrophe (“Parkinson’s disease”) in scientific and popular texts alike. This is likewise the case for Alzheimer and Huntington disease. However, neither Alois Alzheimer, nor James Parkinson or George Huntington suffered from (or possessed) the respective illnesses named after them. Instead they are simply honored as having been the first to describe them. This is different from “Louis Gehrig’s disease” an alternative name used for Amyotrophic Lateral sclerosis or ALS, which became famous as the American baseball player Lou Gehrig actually suffered and died from this illness.
Diseases of the Nervous System. https://doi.org/10.1016/B978-0-12-821228-8.00005-6
109
Copyright © 2021 Elsevier Inc. All rights reserved.
110
5. PARKINSON DISEASE
after-dinner Christmas carols, her mind eased as she had a glass of wine. The next morning, picking up her toothbrush, the shaking started again. In fact, it was out of control. She could not stop the hand from trembling until she held on to the towel rod. Something was terribly wrong. The following day she called her internist, Dr. Helmstedt, who was willing to work her into his schedule that same day. She had not seen him since going off to college. He was kind and caring, yet his questions really irritated her. Had she been using drugs or drinking excessively? He should know her better! As a member of the same church, he knows that she has been teaching Bible school since her junior year in high school. She was not that kind of girl. However, he did not appear to have an answer and told her to keep an eye on it and see if the symptoms wore off. Wear off they did not! The tremors became a daily routine without any sign of abating. Two days later, she passed out in the kitchen without warning while preparing a salad. Yet as quickly as it happened, she regained her consciousness. Her mother, visibly upset, drove her straight to the emergency room at St. Vincent’s Hospital. The waiting room was filled with patients, many in pain, some bandaged from work accidents. She seemed a misfit in their company. After sharing her experience with the triage nurse, she was sent to Neurology for a physical exam, which revealed nothing abnormal. “We need to scan your brain primarily to rule out the obvious,” she was told. What was the obvious? “A tumor or multiple sclerosis,” replied Dr. Nicholas, the neurologist on service. A tumor? What a frightening thought. The MRI machine was making pounding noises and the procedure appeared to last forever. The results of the exam, however, would not be in for a couple of days. Every time the phone rang Anna was startled, afraid to learn she had a brain tumor. When the call finally came, the answer was a great relief. Everything looks normal, with no evidence of any visible brain abnormality. However, Dr. Nicholas explained that he had called in a prescription for a drug called Sinemet, just as an experiment, he said, to rule out Parkinson Disease (PD). “Parkinson’s? Isn’t that what old people get? I am 23 years old!” “Yes, Anna, it is very unlikely,” the neurologist replied. “That is why we are trying the drug only for 10 days as an experiment.” Fortunately, or unfortunately, the drug did the trick. The tremors went away and stayed away. A DNA test since confirmed that Anna carries a rare mutation in a gene called Parkin. As bad luck would have it, Anna was among the very youngest patients to develop this rare genetic form of young-onset Parkinson. Why her? Just the thought of it made her cry. She had just accepted a new job at a publishing firm for children’s books. Would she need to let her new employer know? What was life going to be like? Would she be able to eventually start a family, have children? She will turn 24 tomorrow. She should be living a carefree life, going out with friends, dancing, and having a good time. Yet instead, every time she sees a
clock, she sees her life unfold, imagining her body unable to follow her command, a healthy brain trapped in an immobile body. She was scared.
2 HISTORY In 1817, the British physician James Parkinson published an article entitled “An Essay on the Shaking Palsy,” in which he described the course of disease of six patients who suffered from resting tremors (shaking palsy), abnormal posture, gait difficulties, and reduced muscle strength. He observed three of these patients in the streets of London and the other three in his medical practice. In 1872, a more comprehensive description was compiled by the French physician Jean-Martin Charcot, who studied hundreds of patients to develop a detailed description of the disease. Charcot’s description distinguished the symptoms from the tremors that are often found in patients suffering from multiple sclerosis. He also advocated that the shaking palsy be named PD in honor of James Parkinson’s original description. Charcot already noted that the “muscle weakness” in PD was not caused by a defect in muscle function, but instead was centered in the brain. This notion was confirmed through subsequent autopsy studies by Tetiakoff (1919) and Brissaud (1925), who identified lesions in the substantia nigra (SN) of the midbrain. Their relationship to movement control was proposed in 1938 by Hassler. The proteinaceous inclusion bodies found abundantly in PD brains were independently shown much earlier by the pathologist Frederick Lewy in 1912. These are now the pathological hallmark of PD and are largely composed of misfolded α-synuclein, discovered 80 years later. Interestingly, we still do not know whether these inclusions are the cause or consequence of neuronal cell death. The major turning point in our understanding of PD came in the 1950s. First, it was recognized that dopamine is not merely a precursor for the synthesis of other catecholamine neurotransmitters, such as adrenaline and noradrenaline, but that it was itself a neurotransmitter that was present in significantly high concentrations in the midbrain. Next, the Swedish physician-scientist, Arvid Carlsson, showed that blocking dopamine uptake with the drug reserpine produced Parkinson-like features in rabbits. He was able to reverse the loss of movement in these animals by giving them the dopamine precursor l-DOPA. In 1960, Oleh Hornkiewicz showed a depletion of dopamine in the SN of PD patients. These studies together suggested that a loss of dopamine may be responsible for the motor symptoms, and paved the way for the highly effective dopamine replacement therapy to treat PD patients today. The first human clinical trial was followed soon thereafter and was remarkably successful: “Bed-ridden patients who were unable to sit up, patients who could not stand up when seated, and patients who, when standing, could not start walking
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
3 Clinical Presentation, Diagnosis, and Epidemiology
erformed all these activities with ease after l-DOPA.”1 This p treatment was approved under the name levodopa for clinical use to treat PD in 1967. However, given the systemic conversion of l-DOPA to dopamine by aromatic l-amino-acid decarboxylase, l-DOPA had significant side effects along with short-lived positive results. A more refined delivery regimen was developed by George Cotzias that led to a gradual administration of therapeutic doses of l-DOPA that stabilized the patient on an acceptable dose with remarkable remission of symptoms. However, this was quickly superseded by the discovery that the coadministration of decarboxylase inhibitor carbidopa prevented the systemic conversion of l-DOPA to dopamine, thereby reducing peripheral dopamine side effects and ensuring effective delivery to the brain. The combination of l-DOPA and carbidopa, now in a combination tablet, has since become the standard treatment, approved under the name Sinemet. Many of the contributors to this remarkable discovery were honored with prestigious science prizes. Carlsson was awarded the Nobel Prize in Medicine in 2000, Hornkiewicz received the Wolf Prize in Medicine, and Cotzias was awarded the Lasker Prize. In a historical context, it is interesting to consider the evolution of treatments and the accidental successes achieved along the way. James Parkinson remained committed to the humoral theory prevalent in the day, and therefore suggested blood-letting at the neck as treatment, combined with deliberately placed infections under the skin to divert blood flow away from the brain in order to decompress the brain. By contrast, just a few decades later, Charcot was already experimenting with plant-derived anticholinergic drugs and ergot-contaminated rye grass. The ergot fungi produce ergotamine, which is a synthetic precursor for the synthesis of the modern dopamine agonists, pergolide and cabergoline, used to treat PD. So unbeknownst to Charcot, some of his patients benefitted from the same dopamine agonist frequently used as a first line of treatment today. This history is quite remarkable,2 since this was long before dopamine was even discovered, let alone considered as a neurotransmitter. Equally coincidental, and even more stunning, l-DOPA replacement therapy may have been in use since ancient times in traditional East Indian medicine. l-DOPA is abundantly found in the beans of the cowitch, a tropical legume that is a major ingredient in the traditional Indian medicine Masabaldi Pacana.2 Unfortunately, no specific reports on the use of this remedy for PD can be found in the literature. The dopamine re-uptake inhibitor amantadine may also be added to the list of accidentally discovered drugs. This antiviral agent was used widely in nursing homes to treat infections in the 1960s. It reduced tremors, balance-related issues, and akinesia in Parkinson patients, thereby revealing an unexpected dopaminergic effect. While society is actively discussing the potential benefit of medical marijuana for the treatment of various
111
iseases, it may be of interest that cannabis, either alone d or in combination with opium, was explored as treatment for PD well over a century ago by the British physician Gowers, who reported to have witnessed in his patients a “very distinct improvement for a considerable time under their use” (Gowers, 1899, as discussed in Ref. 2). As is often the case, progress is often propelled by a public fascination with celebrities suffering from disease and who become advocates for the cause. Among the notable personalities who have been living with PD are boxer Muhammad Ali, Olympic cyclist Davis Phinney, and famous actor Michael J. Fox. They were recently joined by rock star Linda Ronstadt and Texas A&M basketball coach Billy Kennedy. The Michael J. Fox Foundation has become a major funding agency supporting PD research.
3 CLINICAL PRESENTATION, DIAGNOSIS, AND EPIDEMIOLOGY 3.1 Epidemiology and Risk Factors Next to Alzheimer dementia, PD is the second-leading neurodegenerative disorder, affecting one million people in the United States and over six million worldwide. Incidence and deaths (~ 210,000/globally per year) have more than doubled in the last 25 years making PD the fastest growing neurological disorder worldwide. This growth is not simply accounted for by an increase in the number of old people, but instead suggest greater exposure to environmental risk factors such as industrial pollutants, chemicals, pesticides, solvents, and metals. Consistent with this notion, highest PD incidence, adjusted for age, is found in the richest, most industrialized countries including the United States. In China, for example, the prevalence of PD doubled in the past 25 years as the country underwent rapid industrial growth. Irrespective of race, PD is 1.4-fold more common in men than in women, and whites are more likely to be diagnosed with PD than are African Americans or Hispanics. Typical disease onset is around 65 years of age, but in rare cases, disease can occur in the 20s or even earlier. Disease incidence increases with age from about 1% of the population at age 65 to 3% at age 85. The vast majority, > 99%, of cases are idiopathic and therefore without a known cause. Rare familial forms are characterized by mutations in a group of genes involved with protein biosynthesis and degradation that include α-synuclein, LRRK2, Parkin, Pink1, and others. The largest risk factor for PD is age; however, environmental factors are suspected to play an important role in the disease etiology. People growing up in rural areas are more likely to develop PD, and exposure to pesticides such as rotenone, or herbicides such as Agent Orange, is suspected risk factors. Surprisingly, smoking cigarettes or using nicotine or caffeine reduces the
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
112
5. PARKINSON DISEASE
risk to develop PD. The discovery that MPTP (1-methyl-4phenyl-1,2,3,6-tetrahydropyridine), a by product of trying to illicitly manufacture heroin, produces a remarkably similar disease phenotype has spurred interest in identifying related environmental toxins that may be causal for the disease. However, while MPTP provides one of the most widely used animal models to study PD, no related chemicals have yet surfaced that may cause idiopathic PD. Nevertheless, current evidence strongly suggests that environmental factors play a major contributing role in the disease, probably acting synergistically with an inherent genetic risk.
3.2 Disease Presentation and Diagnosis PD is characterized by a group of “cardinal” features that include rigidity, resting tremor, slowness in movement (bradykinesia), and balance problems. These symptoms are usually asymmetric, affecting one side of the body more than the other. Tremor is an early common symptom that is easily recognizable even by a layperson. It initially affects only one body side and is most noticeable when the affected limb is at rest, and disappears with purposeful movement. There is a smooth quality to the tremor, as if a person were rolling a marble between thumb and index finger. Hands and arms are more affected by the tremor than the lower extremities, and tremors can become quite bothersome for the patient, preventing skilled fine motor tasks, holding a glass without spilling its contents, or placing a key into a lock. There is also a pronounced overall slowness (or poverty) of motion, called bradykinesia. This impairs the initiation of movement and makes individuals appear frozen. This goes along with shortened stride length, decreased arm swings, and lower rate of eye blinks. The stoic appearance is further enhanced by overall rigidity and stiffness. Once walking, a person with PD is hunched forward and walks with a slow, narrow, shuffling gate (Figure 1). As the disease progresses, the patient becomes unsteady and falls frequently. The major defect that explains the impaired motor symptoms is a loss of dopamine-producing neurons in the substantia nigra (SN) pars compacta, a nucleus in the midbrain that is part of the basal ganglia. This structure communicates with the motor cortex and the spinal cord and is involved in the modulation of movement discussed in greater detail in Section 4.1. Restoration of dopamine by administration of the dopamine precursor levodopa transiently restores normal motor function and serves as an important tool to unequivocally diagnose PD and to differentiate it from other, related movement disorders. In addition to neuronal loss in the basal ganglia, nondopaminergic neurons are lost in other parts of the brain. Recent studies actually suggest that the disease may begin in the peripheral autonomic nervous system, spreading centrally over time.3 This may explain the early loss of olfaction, which is a telltale sign that often precedes disease onset by several years.
FIGURE 1 Front and side views of a man with a festinating or f orward-leaning gait characteristic of Parkinson Disease. Drawing after St. Leger, first published in Wm. Richard Gowers’ Diseases of the Nervous System, in London, 1886 (public domain).
A diffuse loss of nondopaminergic cells may also underlie the progressive development of dementia, which eventually affects 80% of patients with PD. Given that dopamine plays a role in the reward system that controls mood and sexual and pleasurable behavior, many PD patients show behavioral changes that may include depression and alterations in sex drive. Other rare psychiatric problems in PD include compulsive gambling, particularly when PD patients are treated with dopaminergic drugs.
3.3 Disease Stages It is common to categorize four stages of PD, which differ among patients in severity of symptoms and the degree of impairment. In the premotor stages of the disease, constipation, loss of olfaction, and abnormal sleep, particularly the acting out of dreams, may predict the future onset of disease. Ninety percent of PD patients have a measurably decreased sense of smell at time of diagnosis, and a recent prospective study showed that a poor performance on olfactory tests was highly predictive of developing PD years later.4 Another study of 6790 men without PD showed that those with infrequent bowl movements had a fourfold increased risk to develop PD.5 Other symptoms that may precede diagnosis by many years include restless leg syndrome, anxiety, and changes in cognitive function. The latter may include subtle attention problems and changes in executive function, including problems with planning, abstract thinking, and cognitive flexibility. Early stages of PD are dominated by the cardinal motor symptoms, that is, resting tremor, postural i nstability, bradykinesia, and rigidity. At this stage, levodopa is
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
3 Clinical Presentation, Diagnosis, and Epidemiology
aximally effective, and in many patients, the drug m controls motor symptoms so entirely that the patient has little impairment at work or in social life. The “off” states, which are the times that the medication has worn off, are short, and repeated dosing of levodopa maintains the patient in a well-functioning and “on” state all the time. Over time, the period during which complete dopamine control can be achieved grows shorter. As patients enter a period of moderate PD, they spend more time in the disabled “off” state. They may also begin to experience short periods of dyskinesia, although the impairments are not at the point that the patient would wish to discontinue levodopa treatment. Constipation becomes progressively worse, and depression is seen in close to 50% of patients at this stage. In advanced stages of disease, motor and nonmotor symptoms are quite disabling as the levodopa effects are short lived and incomplete. Gait problems worsen, resulting in instability and frequent falls. Dementia and behavioral problems become significant and even in cases where the use of a deep brain stimulator ameliorates motor symptoms, nonmotor symptoms are a major disability, and the burden on caregivers and family members is tremendous. Histopathologically, the neuronal loss in the substantia nigra can readily be seen on autopsy, where the black melanin-pigmented neurons that give the SN its name are missing in PD patients (Figure 2). In addition, protein aggregates called Lewy bodies that contain α-synuclein are found diffusely throughout the brain (Figure 3). New imaging techniques allow visualization of the early dopaminergic neuronal loss by positron emission tomography (PET) (Figure 4), taking advantage of the fact that these neurons bind the 18F-labeled dopamine precursor l-DOPA (fluorodopa). This approach, however, is not commonly used clinically. There is currently no cure for PD, but a number of disease-modifying treatments are available that potentially slow the progression of disease and greatly improve
113
FIGURE 3 Lewy body pathology. A high-resolution histological image of the substantia nigra shows neurons containing brown melatonin granules along with characteristic Lewy bodies (arrow). Image was kindly provided by Dr. Thomas Caceci.
the quality of life for most patients. These are largely targeted at the motor symptoms and seek to restore normal dopamine levels in the brain. This is primarily accomplished through dopamine replacement therapy, which involves the systemic administration of dopamine precursor levodopa in conjunction with enzyme inhibitors that reduce peripheral degradation of this prodrug. Unfortunately, the effectiveness of this strategy wears off over a period of 5–10 years and yields somewhat abnormal hyperkinetic, jerky movements known as “chorea,” akin to that seen in Huntington Disease. Recent successes using implanted electrodes that stimulate motor control pathways in the midbrain have provided significant benefit to many patients, and this deep brain stimulation (DBS) is now widely used in advanced stages of PD. Many of the nonmotor effects of PD cannot be effectively treated. Of these, a somewhat unique form of dementia, which primarily affects executive functions and planning of daily activities, but without memory loss, is a major impairment for which no effective treatment
FIGURE 2 Loss of pigmented dopamine neurons in the substantia nigra from a patient with PD (left) compared to a normal substantia nigra from a healthy individual (right). The distinct loss of the black-appearing dopaminergic cells that contain the black pigment neuromelanin, serving as a marker for dopaminergic neurons, is prominently seen in the substantia nigra from the patient with PD. Reproduced with permission from Ref. 6.
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
114
5. PARKINSON DISEASE
FIGURE 4 Presence of dopamine neurons assessed by 18-fluorodopa positron emission tomography (PET). (A) PET scan from a control subjects showing high striatal uptake of the precursor used to synthesize dopamine, indicative of the presence of dopaminergic neurons (highest value in red). (B) Example of a patient with Parkinson Disease with motor signs mainly confined to the left limbs. Uptake of the dopamine precursor is markedly reduced in the right posterior putamen (area indicated by arrow is 70% below normal) and to a lesser extent in the anterior putamen and caudate of the left hemisphere. Reproduced with permission in a modified form from Ref. 7.
exists. In addition, depression is very common, affecting between 50% and 80% of PD patients, and often goes untreated. With disease progression, the strain on the caregiver becomes significant, and round-the-clock residential nursing may be advisable, if possible. Most PD patients live to an almost-normal life expectancy. The early phase of disease, during which a patient is able to go about his or her daily life with minimal motor deficits, lasts for about 5 years, during which time period most patients can maintain employment. This is followed by a 5–10-year period during which dopamine replacement therapy provides sufficient relief to manage the disease, albeit with impairments. An additional 10 years of quality life may be gained in some patients through implantation of a deep brain stimulator. Although PD is typically considered a nongenetic disorder, about 15% of patients have a first-degree family member who also suffers from PD. In about 5% of these, de novo mutations in the LRRK2 gene are found, but causality to disease for these mutations is rarely proven.
3.4 Related Movement Disorders Essential tremor (ET): Although tremor is one of the hallmarks of PD, tremors also occur in other movement disorders and, indeed, we all have physiological tremors that are quite normal. Normal tremors are typically invisible, but can show up in periods of stress or after consumption of coffee or nicotine. One of the diseases often misdiagnosed as PD is called ET or benign tremor. It affects primarily the arms and hands, but can include the head and even voice. These tremors are high-frequency movements (4–12 Hz) that begin unilaterally and, with
FIGURE 5 “Archimedean” spiral drawn by an individual with tremor (B) and an unaffected individual without tremor (A). The jittery appearance of the spiral indicates the presence of tremor. Reproduced with permission from Ref. 8.
time, involve both sides of the body. A simple way of diagnosing and monitoring ET is having the patient draw an “Archimedean” spiral (Figure 5), which should look smooth in a normal person but is jagged in a person suffering from ET. Head tremors cause shaking “no–no” or “yes–yes” movements. ET is the most common movement disorder and affects 5% of the general population and up to 9% of the population over age 60. Tremors typically worsen with age, becoming higher in amplitude and therefore more pronounced. These particularly impair the dominant extremities and can make simple tasks such as eating, drinking, or personal hygiene difficult, and operating machinery impossible. Since ET was initially believed to only involve motor function while sparing overall cognitive health, these were originally called benign tremors. However, there is now clear evidence that patients with ET also develop cognitive decline, anxiety, and hearing loss. Although ET does not reduce
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
4 Disease Mechanism/Cause/Basic Science
life expectancy, it can nevertheless be quite disabling. Fortunately, for many patients, tremors are treatable. Arm and hand tremors respond well to simple beta-blockers such as propranolol, with few side effects. Others respond to the phenobarbital analogue primidone. Severe cases may require placement of a deep brain stimulator. Dystonia: Involuntary sustained muscle contractions that cause the body to twist into an abnormal shape, often with spastic contractions, are the classical symptoms of a heterogeneous collection of conditions called dystonias. These symptoms can start in early childhood, and the earlier their onset, the more severe the condition becomes. Among adults, cervical dystonia is the most common, with 15,000 newly diagnosed patients each year in the United States. It causes the neck to be unusually tilted or rotated and is a very painful condition. Dystonias can be the primary presenting feature, or they can be secondary to an injury or disease. The recent identification of over 20 genes involved in dystonia suggests that there is a clear underlying genetic cause for most forms of dystonia. In spite of the advances in identifying disease-causing genes, little is known about its cause, and treatment options are few and largely ineffective. In rare forms, patients respond to levodopa; in others, anticholinergic drugs or GABA agonists such as baclofen or clonazepam may help. If the dystonia is focal, botulinum toxin injections can transiently relieve symptoms. DBS is also being used in severe forms of dystonia, with varying success.
115
FIGURE 6 The motor homunculus based on Wilder Penfield’s original drawings from 1940 shows the cortical representation of our body with regard to control of voluntary movement (public domain).
4 DISEASE MECHANISM/CAUSE/BASIC SCIENCE Since PD is by far the most studied and best understood of the movement disorders, we will focus almost exclusively on its etiology. Because it is primarily a movement disorder, it is useful to briefly review the functional neuroanatomy of normal control of movement, as this will aid our understanding of the defects present in patients with PD.
4.1 Dopamine and the Control of Movement Willful movements originate from the primary motor cortex, which is located anterior to the central sulcus. Similar to the sensory cortex, the motor cortex is organized somatotopically, and this organization is called the motor “homunculus” (Figure 6). The innermost region of the cortex controls movements of the foot, and the outermost regions control hand, face, and tongue. Given the importance of fine movements in our hands, made possible through coordination of 34 distinct muscles and 123 ligaments, a disproportionately large amount of motor cortex is devoted to it. There are several parallel pathways projecting from the primary motor cortex to the body, with some
FIGURE 7 Cortical control of movement involves primary motor cortex and premotor and supplementary motor areas. Their activity is modulated by the midbrain, particularly the basal ganglia. Reproduced with permission from Ref. 9.
redundancy between them. Direct projections via the corticospinal tract innervate motor neurons in the spinal cord that control muscles in the extremities. Stimulation of neurons with such direct projections in the primary motor cortex causes discrete movements that involve only a small muscle group. An example would be bending of the index finger. The vast majority of movements, however, arise differently. For these, the primary motor cortex activates adjacent cortical premotor and supplementary motor areas (Figure 7). Stimulation of individual neurons in these associated cortical regions elicits
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
116
5. PARKINSON DISEASE
more complex patterns of movement, such as gestures or poses that require coordinated activation of a number of muscle groups. Examples would be a leg moving in a walking stride or an arm folding over the head. These patterned movements involve projections from the primary motor cortex via the basal ganglia back to the premotor and supplementary motor cortex (Figure 7).
4.2 The Basal Ganglia The basal ganglia, illustrated in Figure 8, are a collection of nuclei at the interface of the telencephalon, diencephalon, and mesencephalon. In the telencephalon, four structures collectively form the corpus striatum, which we will simply call striatum. These are the caudate nucleus, putamen, nucleus accumbens, and globus pallidus. In the diencephalon resides the subthalamic nucleus (STN), and the single mesencephalic nucleus is called the SN. It has two parts, the pars compacta and pars reticulata. Loss of dopaminergic neurons in the pars compacta causes PD, as already illustrated in Figure 2. The major input to the basal ganglia comes from the cortex and terminates in the striatum, which runs just underneath the cortex along its entire length. Like the cortex, the input to the striatum is topographically organized, with frontal cortex innervating frontal striatum and posterior cortex innervating posterior striatal
eurons. The major output structures from the basal n ganglia are the globus pallidus interna and the SN pars reticulata. These signal to the thalamus, which in turn relays this information to the associated motor cortex. The projections from the cortex to the striatum are excitatory and use Glu, while the output from the basal ganglia to the thalamus is GABAergic and therefore inhibitory. The basal ganglia integrate motor information to filter and integrate incoming and outgoing signals. Overall, their function would be best described as a brake on the motor output; they provide a tonic inhibitory influence on the motor output from the premotor and supplemental motor cortex. For any motor activity to occur, the brake must be transiently removed. This is achieved through dopamine release from neurons in the substantia nigra, which causes a disinhibition of the output from basal ganglia projecting back to the cortex. High levels of dopamine therefore promote movement (reduced brake action), while low levels promote inactivity (increased brake action). This explains the slowness of movement when dopamine is lost in PD and the chorea-like hyperactive movements when a PD patient has elevated dopamine. To understand this schema in more detail, we will next examine the pathways and projections involved. There are two distinct pathways at work that balance each other’s activity. These are called the direct and indirect pathways, schematically illustrated in Figure 9.
FIGURE 8 The basal ganglia include the caudate nucleus (CN), putamen (Put), nucleus accumbens (Acb), and globus pallidus (GP). All four are part of the telencephalon. The subthalamic nucleus (STN) and substantia nigra (SN), with its two parts, pars compacta (SNc) and pars reticulata (SNr), modulate the output of the globus pallidus to the thalamus to control the associated motor cortex. Produced using the 3D Brain App, Cold Spring Harbor Laboratory DNA Learning Center.
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
4 Disease Mechanism/Cause/Basic Science
Direct pathway
Indirect pathway Cortex
+ Glu
SNc
+
+
DA
+ Glu
Striatum
ACh
–
+
SNc DA
ACh –
GABA, Enk
GPe
GABA –
GABA – SP
Glu
STN
+
GPi SNr
GPi SNr
– GABA
– GABA
Thalamus (VA, VL) +
Glu + Glu To cortical motor areas
FIGURE 9 Direct and indirect pathways of movement control by the basal ganglia illustrated as a schematic. Abbreviations are as follows: glutamate—Glu; acetylcholine—ACh; enkephalin—Enk; dopamine—DA; substantia nigra pars compacta—SNc; substantia nigra pars reticularis—SNr; globus pallidus external—GP3; globus pallidus internal—GPi; subthalamic nucleus—STN. Reproduced with permission from Ref. 10.
The direct pathway sends inhibitory projections from the striatum to the globus pallidus internal (GPi). The GPi then sends inhibitory projections to the thalamus, which sends excitatory projections to the associated motor cortex. If we add this all up, activity of the direct pathway causes excitation. One easy way to remember this is to simply multiply the excitatory (+ 1) and inhibitory (− 1) activities. We have two inhibitory and one excitatory:
1 1 1 1 excitatory action on the cortex. The indirect pathway starts in a different set of neurons in the striatum that sends inhibitory projections to the globus pallidus external (GPe). These neurons make inhibitory projections to the STN, which in turn make excitatory connections to the GPi. This in turn is connected, as we learned earlier, by inhibitory connections to the thalamus, which then excites the motor cortex. Thus, for the net effect, we have three inhibitory and one excitatory connection:
1 1 1 1 1 inhibitory action on the cortex. Both of these pathways are under the direct control of the dopaminergic output from the SN that project to the striatum, called the nigrostriatal projections. The striatal neurons that receive synaptic input from the striatum do so via D1 dopamine receptors, which depolarize in
117
response to dopamine, resulting in excitation. By contrast, the striatal neurons that originate the indirect pathway express D2 dopamine receptors that hyperpolarize the cell in response to the same input. Both act reciprocally, with increases in dopamine activating the “excitatory” direct pathway while inhibiting the “inhibitory” indirect pathway. In PD, the nigrostriatal projections are lost, which causes the excitatory direct pathway to be silent and the inhibitory indirect pathway to be active, resulting in reduced motion. The above-described parallel loops of direct and indirect pathways that complement each other’s activity to assure smooth motor control were described in the 1980s and have served us well toward understanding the pathophysiology of movement disorders. However, although this schema already appears complicated, it does not nearly capture the complexity of the actual wiring of the basal ganglia, and leaves out many projections that contribute to the integrative functions that these nuclei perform. Even today, much of basal ganglia function remains only partially understood. It must be stressed that the above model only explains one of the cardinal motor features of PD, namely, a poverty of spontaneous movement, or bradykinesia. It does not explain the rigidity and tremor at all. Moreover, it also provides little toward our understanding of the nonmotor symptoms.
4.3 Lewy Bodies Pathology and Progression of Disease As with Alzheimer and Huntington Disease, the brain of patients with PD shows intracellular protein aggregates. These were already recognized in 1912 by Friedrich Lewy, and hence named Lewy bodies and are the pathological signature for PD (Figure 3). They are formed predominantly by accumulations of α-synuclein and fill the cell bodies (perikarya) and processes of neurons largely in the CNS and to a minor extent also in neurons of the enteric nervous system. They are sometimes found in oligodendrocytes as well, albeit much less frequently. Lewy bodies contain dense globular material as well as 10–15-nm straight filaments, which can be formed from recombinant α-synuclein in the test-tube. α-Synuclein is an abundant protein highly expressed in neurons of all types. It accounts for 1% of total cytosolic protein in the nervous system; however, its function is not entirely understood. It is found in close proximity to synaptic vesicles and appears to modulate the SNARE complex of proteins (SNARE—Soluble NSF Attachment Protein, NSF—N-ethylmaleimide-sensitive factor) involved in vesicle binding, having a dampening effect on synaptic vesicle release. Cytosolic α-synuclein does not have a tertiary structure, yet when it associates with phospholipid membranes, it assumes an α-helical
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
118
5. PARKINSON DISEASE
s tructure. When it is overexpressed, however, it changes to a β-sheet formation, which, much as is the case with amyloid in Alzheimer Disease, has a tendency to assemble into sticky aggregates. In PD, mutations in the α-synuclein gene cause a doubling or even tripling of the encoded gene, leading the cell to make an overabundance of the protein. Moreover, overexpression and abnormal phosphorylation, particularly at serine residue
129 (pSer129-α-synuclein), or modifications by reactive oxygen species contribute to the filamentous aggregation of α-synuclein into Lewy bodies. In addition to enhanced production, inadequate protein degradation via the ubiquitin-proteasome pathway (Box 1, UPS) likely contributes to Lewy body formation as well. Lewy bodies are immune-reactive for ubiquitin and ubiquitin-binding proteins.
BOX 1
THE UBIQU ITIN-PROTEASOME SYSTEM (UPS). Most nervous system proteins are short lived (hours to days) and are being constantly replaced. We refer to this as protein turnover. Also, misfolded or otherwise damaged proteins are rapidly cleared from the cell. Protein degradation and recycling occurs via the ubiquitin-proteasome system (Box 1-Figure 1). At the heart of this machinery is the proteasome, a large intracellular complex that binds and enzymatically cleaves proteins through proteolysis. It works in conjunction with a family of enzymes that targets proteins for degradation by attaching ubiquitin labels to their surface. Ubiquitin is a small, abundant 76-amino-acid polypeptide that becomes activated by three enzymes called E1/E2 and E3 ubiquitin ligases. By using ATP, these activate and then covalently bind ubiquitin to a protein substrate. Multiple ubiquitins are typically attached. The
result is polyubiquitinated protein recognized by the 19S proteasome cap, which removes the ubiquitin for future use. It then shuttles the protein to be degraded into the hollow core of the proteasome, where the protein is cleaved into its individual amino acids. Here a number of enzymes work together, resembling the activity of caspase, trypsin, and chymotrypsin, to cleave the protein. The identification of which proteins are to be targeted for degradation rests entirely on the ubiquitin ligases, particularly the E3 ligase, the last one in the cascade. A number of proteins have been shown to harbor this activity. Notably, Parkin, which is one of the mutated proteins in familial PD, has been identified to harbor E3-ligase activity. Mutated Parkin may therefore cause aberrant protein turnover that may contribute to the formation of Lewy bodies.
BOX 1-FIGURE 1 The ubiquitin-proteasome pathway. Target proteins of the proteasome are tagged with polyubiquitin molecules in an ATP-dependent process through E1, E2, and E3 ligases. Polyubiquitinated proteins are then recognized by the 19S regulatory complex of the 26S proteasome and fed into the 20S catalytic core for degradation and the ubiquitin molecules recycled. Reproduced with permission from Ref. 11.
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
4 Disease Mechanism/Cause/Basic Science
Whether Lewy bodies are directly contributing to disease pathology or are simply an inevitable aggregation of diseased proteins, possibly as a protective measure, is not entirely clear. However, recent studies make a strong case for Lewy bodies or at least the α-synuclein contained in them being disease causing. Specifically, examination of human postmortem brains from individuals who died at various clinical stages of PD shows a stereotypic spread of α-synuclein throughout the brain, whereby pSer129-α-synuclein pathology first appears in the olfactory bulb, followed by the motor nucleus of the vagus nerve, then the medulla. Only later in disease progression does α-synuclein populate the midbrain and the substantia nigra, and eventually cortical brain regions throughout the brain.3 Once seeded, the spread of α-synuclein oligomers occur in a prion-like fashion, similar to what we discussed for β-amyloid in AD, whereby the misfolded α-synuclein spreads from cell to cell and coaxes wild-type α-synuclein into its misfolded shape. Interestingly, the diseases is now believed to start peripherally in the gut, with misfolded α-synuclein being transported and delivered to the brain via the vagus nerve that innervates the ileum.12 A harmful effect of α-synuclein on enteric nerves and/or muscles of the ileum may explain why constipation often precedes the disease onset by many years, and why the first CNS symptoms are loss of olfaction, as the olfactory bulb being among the first brain regions to be populated with misfolded α-synuclein. Note that subsequent spread of pathology throughout the cortex explains why PD eventually affects nonmotor
119
functions as well, causing cognitive decline, emotional abnormalities, dementia, and depression.
4.4 How Might α-Synuclein Aggregation in Lewy Bodies Contribute to Disease? Little is known concerning the function of normal α-synuclein. Given that it accounts for 1% of a neurons protein, it has to be important. Lewy bodies do not kill neurons, nor is the presence of phosphorylated or mutated α-synuclein toxic. Transgenic mice with α-synuclein deletion exhibit increased neurotransmitter release, suggesting that α-synuclein dampens synaptic function.13 Recent studies show a direct interaction of α-synuclein with molecules that form the SNARE complex responsible for orchestrating presynaptic vesicle docking, fusion, and transmitter release.14 Synuclein binds directly to synaptobrevin, also known as vesicle-associated membrane protein (VAMP), and stimulates the assembly of a synaptic vesicle complex (Figure 10). Synuclein serves as a molecular chaperone that regulates the availability of synaptic vesicles for release, a function that is impaired when α-synuclein is mutated or aggregated into Lewy bodies. In light of this, one could conclude that Lewy bodies, by aggregating α-synuclein, deprive the cell of sufficient α-synuclein to support normal synaptic vesicle release, which, dependent on the affected neuron, may impair contractions of the gut, reduce olfaction, and impair dopamine release in the striatum.
FIGURE 10 Role of synuclein in synaptic function. Synuclein acts as a molecular chaperone, directly interacting with the vesicle-associated membrane protein (VAMP) that stimulates the assembly of vesicles containing neurotransmitter. Reproduced with permission from Ref. 15.
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
120
5. PARKINSON DISEASE
4.5 How Might We Explain the Progressive Neuronal Loss Primarily in the SN? The most prominent histopathological feature of PD is the selective loss of dopaminergic neurons in the SN. As illustrated from lesion experiments, for example, as seen with exposure of drug addicts to the neurotoxin MPTP, this loss of neurons is sufficient to explain the motor symptoms of PD and it is generally accepted that a loss of these neurons causes PD. That said, this population of neurons is very small when compared to the rest of the brain. What causes these neurons to die while others are spared? If we knew the answer to this question, we might be able to prevent the motor symptoms of PD from occurring. There are a number of plausible vulnerabilities emerging, and some may ultimately lead to novel ways of preventing disease.16 A starting point to answer this question is to look at the major neuronal types in the basal ganglia and the transmitters they use. The striatum is by far the largest structure within the basal ganglia. It runs the entire length of the cortex, where it receives input from both ipsilateral and contralateral cortex. The primary striatal neuron is called the medium spiny neuron; this type of neuron makes up 95% of all neurons in the striatum. These cells have elaborate dendritic arbors that are studded with dendritic spines (Figure 11). Spiny synapses are always excitatory and are the sites of glutamatergic input from the cortex. These neurons also receive dopaminergic input from the SN via the nigrostriatal pathway. Neurons originating in the substantia nigra have extensively branched axons, with each making up to one million synaptic contacts onto striatal neurons, where they primarily release dopamine.
The two pathways involved in the control of movement not only differ with regard to their output but also receive their dopaminergic input via different dopamine receptors. The direct pathway responds via D1 receptors; the indirect pathway responds via D2 receptors. Both are G-protein-coupled receptors (GPCRs). D1 receptors couple via Gs-type proteins to activate adenylyl cyclase to increase the production of cAMP. The D2 receptors do the opposite. They couple via Gi proteins to inhibit adenylyl cyclase and therefore inhibit production of cAMP (Figure 12). The concentration of cAMP, in turn, changes the excitability of striatal neurons through complex interactions with ion channels and transmitter receptors. D1 activation causes depolarization and enhances excitability; D2 activation is inhibitory. There are equal numbers of D1-expressing and D2-expressing medium spiny neurons in the striatum. The release of dopamine occurs from GABAergic neurons in the SN. These neurons have intrinsic rhythmic electrical activity and function as “pacemakers” that drive tonic dopamine release in the striatum. Rhythmic activity entails frequent depolarization, during which Ca2 + enters cells through voltage-activated Ca2 + channels. This Ca2 + entry is an integral part of the pacemaker activity, as it activates Ca2 +-activated K+ channels required for the repolarization following each Na+dependent depolarization. This constant influx of Ca2 + is one vulnerability believed to play an important role in the pathogenicity in PD. For the pacemaker neurons to fire action potentials (APs) at a rate of 5–10 per second, they require a lot of energy. Each AP causes a Na+ influx that needs to be removed via the Na+/K+-ATPase. Each membrane depolarization also activates Ca2 + channels, causing a Ca2 + influx. Unlike many other pacemaker neurons, SN neurons do not have significant Ca2 + buffering capacity in the form of proteins like parvalbumin that bind Ca2 +.
FIGURE 11 A medium spiny neuron in the mouse striatum
FIGURE 12 Stimulation of the D1 receptor (left) couples via the
densely covered with synaptic spines. These are the sites receiving glutamatergic input. The cell was filled with biocytin and labeled with a streptavidin-conjugated fluorophore (green). Image was kindly provided by Dr. Rita Cowell, University of Alabama Birmingham.
stimulatory G protein Gs to increase cAMP, causing the striatal neuron to be excited, whereas activation of the D2 receptor (right) couples via the inhibitory G protein Gi to reduce cAMP levels and inhibits striatal neurons. National Institutes on Drug Abuse.
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
4 Disease Mechanism/Cause/Basic Science
Therefore, Ca2 + must be constantly shuttled out of the cell or into organelles against a steep (10,000-fold) concentration gradient, again taxing cellular energy. As illustrated in Figure 13, a number of pump systems that are either directly or indirectly fueled by ATP are constantly at work to maintain this balance. The Na+/K+ ATPase returns the Na+ and K+ concentrations to normal. Multiple Ca2 + ATPases remove the Ca2 + that enters as cells depolarize. The synaptic vesicles are loaded using a H+ gradient across the vesicle that is established by a vacuolar H+-ATPase. To meet this incredible energy demand, SN neurons must use every ounce of glucose to produce ATP in the mitochondria by oxidative metabolism, which generates 36 moles of ATP per mole glucose. This conversion occurs in the cellular energy generators, the mitochondria. Not surprisingly, SN neurons are rich in mitochondria. To be particularly effective in providing energy where it is most needed, mitochondria travel close to the sites of highest energy demand: the axon initial segment, where the rhythmic APs are generated, and synaptic terminals, where energy is needed to sustain synaptic activity. The emerging picture is that SN neurons require very high mitochondrial ATP production and, as such, are dependent on superior mitochondrial function and health. Of course, we already learned that essentially all neurons are vulnerable to cell death. This is perhaps best exemplified by ischemic stroke, where energy substrates are lost. Yet the SN neurons have an additional vulnerability with regard to mitochondrial function.
121
Several chemicals that were shown to kill SN neurons and produce Parkinson-like symptoms turned out to be mitochondrial poisons. The pesticide rotenone, the defoliating chemical Agent Orange, and the synthetic heroin by product MPTP are prime examples discussed in more detail later. It is well known that during normal oxidative ATP production, mitochondria are somewhat sloppy in their handling of oxygen and produce some reactive oxygen species (ROS) (Figure 14). These are highly reactive molecules with unpaired electrons and include superoxide O 2 −, hydrogen peroxide H2O2, peroxynitrate ONOO∙, and hydroxyl radicals OH−. These ROS molecules can react with lipid membranes, causing lipid peroxidation, but can also damage RNA and DNA. ROS
FIGURE 14 Mitochondrial production of energy by oxidative phosphorylation produces reactive oxygen (ROS) and nitrogen species (NOS), particularly O 2 − , OH−, and H2O2. These radicals with unpaired electrons are highly reactive and are destructive to DNA, RNA, proteins, and lipids. In PD, the high-energy consumption of pacemaking neurons generates an overabundance of ROS and NOS.
FIGURE 13 Ion flux via numerous ion channels and transporters across the neuronal membrane is essential for neuronal activity. Fluxes are particularly pronounced in pacemaking neurons, which persistently fire action potentials at a high frequency. Ionic movement is directly or indirectly fueled by ATP, putting cells at a high risk for failure should energy supplies deplete. Reproduced with permission from Ref. 16.
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
122
5. PARKINSON DISEASE
damage to mitochondria induces apoptotic cell death and, consequently, enzymes such as superoxide dismutase or cellular antioxidants such as glutathione attempt to contain toxic concentrations of ROS in healthy mitochondria. The mitochondria are unique organelles that contain their own mitochondrial DNA (mDNA), through which they produce many of the enzymes required for oxidative metabolism. Most important among these are proteins that form complexes I–IV, the sites at which the proton gradient across the mitochondrial membrane is converted to chemical energy in the form of ATP. This mDNA is at constant risk of being damaged or mutated through mitochondrially produced ROS. This in turn can result in further production of ROS, causing a vicious cycle whereby mitochondrial ROS causes further damage. To add insult to injury, the oxidation of dopamine and its metabolites can produce additional ROS, and levels of the protective antioxidant glutathione are reduced in PD patients. As we will see further in the section on genetics of PD, mutations that occur in familial forms of PD affect two proteins, Parkin and PINK1. These proteins normally recognize damaged mitochondria and assure that they are disposed of rather than producing further damaging ROS; but, after mutation, these proteins now keep faulty mitochondria hanging around SN cells. Interestingly, a relationship between metabolism, energy use, and aging had already been proposed as long ago as the early 1900s, based on the finding that organisms with low baseline metabolism live longer than those with high energetic demand. Subsequently it was proposed that degenerative diseases may be related to the damaging effects of ROS (Harman, 1956, as discussed in Ref. 16). The oxygen-free radical theory of aging then morphed into the mitochondrial theory of aging, with the recognition that most of the ROSs were produced in conjunction with the mitochondrial electron chain.
4.6 Why Does It Take 60 Years or Longer for PD to Develop? As with all neurodegenerative diseases, the time course of disease is a big question mark. Cumulative toxicity, combined with environmental exposure risks that can hasten the onset of disease, is certainly an attractive explanation for the progressive late-onset nature of PD. While we do not know if any of the above-discussed vulnerabilities occur in dopaminergic SN neurons, one can make the argument that from an evolutionary point of view it is acceptable for these neurons to be vulnerable and “live on the edge.” Note that the safety margin for dopamine in the brain is extraordinary. Approximately 80% of all dopaminergic SN neurons must die before motor symptoms manifest clinically. Obviously, in most of us, an 80% loss does not occur over our lifetime and we do not develop PD. In those who unfortunately do develop PD, it typically will have taken 60 years to develop. This is well past the usual years of procreation for most people, and therefore the continuation of the human race is not at risk because of this neuronal loss. Because of this, there is no evolutionary pressure to select against this phenotype.
4.7 Genetics of PD The vast majority of PD cases show no evidence for heritance and are considered idiopathic or with unknown cause. However, a small but significant number of genetic causes of the disease have been reported and provide a starting point toward a molecular understanding of the underlying disease mechanisms that are likely shared in PD. The first disease-causing genetic mutation in PD was reported in 1996. Today, there are 28 known chromosomal regions that harbor disease-causing mutations, yet only six contain genes that, when mutated, actually cause monogenetic PD17 (Table 1). All six are autosomal, linked to chromosomes other than sex chromosomes. Two are dominant and four are recessive genes. In dominant inheritance, one
TABLE 1 Six genes have been identified as causes of familial PD. Protein
Gene
Heritance
Presentation
Function
α-Synuclein
SCNA
Dominant
Early onset
Involved in synaptic vesicle release
LRRK2 (leucine-rich repeat PARK8 kinase 2)
Dominant
Mid-late onset
Large multifunctional signaling protein; synaptic transmission
Parkin
PARK2
Recessive
Juvenile Parkinson Disease (PD)
E3 ubiquitin ligase
PINK1 (PTEN-induced putative kinase 1)
PARK6
Recessive
Juvenile and early onset
Targets malfunctioning mitochondria
DJ-1
PARK7
Recessive
Early onset
Cytoplasmic sensor for oxidative stress
ATP13A, probable cationtransporting ATPase
ATP13A2
Recessive
Early onset
Lysosomal membrane protein
These encode proteins of different function and characterize different variants of disease as indicated.
II. PROGRESSIVE NEURODEGENERATIVE DISEASES
4 Disease Mechanism/Cause/Basic Science
mutated allele is sufficient to cause disease, and inheritance usually shows affected individuals in every generation. By contrast, recessive genes only become symptomatic if both alleles are mutated, and these typically skip generations. The two autosomal dominant mutations unequivocally linked to the cause of monogenetic PD are α- synuclein and LRRK2. Carriers of mutated α-synuclein present with early-onset PD ( 2.5 cases per 100,000 (8000 cases each year in the United States). After the introduction of an efficacious vaccine against Haemophilus influenzae, the majority of cases (~ 60%) in the United States are now caused by Streptococcus pneumoniae, even though the epidemiology of this infection has been altered by an effective vaccine. The second most common cause for bacterial meningitis (~ 35%) is Neisseria meningitides, also called meningococcus. Its incidence has declined significantly due to routine childhood vaccination with a heptavalent pneumococcal conjugate vaccine and a vaccine requirement for college attendance. Due to the relative absence of routine vaccinations, preventive antimicrobial treatments, and the immune-compromising effect of HIV infection, the number of people, particularly children, suffering from bacterial meningitis is up to 100-fold greater in the developing world where an annual 1.2 Mio new cases present each year. For example, sub-Saharan Africa and surrounding peri-Sahelian countries comprise the African “meningitis belt,” where
III. SECONDARY PROGRESSIVE NEURODEGENERATIVE DISEASES
238
10. Infectious Diseases of the Nervous System
TABLE 1 Examples of CNS infections by infectious agents. Infectious agent
Disease
Pathogen
Treatment
Prevalence (US/global)
Outlook
Bacterium
Meningitis
Streptococcus
Antibiotics
Low/low
Neurosyphilis
Treponema pallidus
Antibiotics
Low/medium
Good to excellent with treatment
Lyme disease
Borellia
Antibiotics
Low
Leprosy
Mycobacterium
Antibiotics
0/Medium
Meningitis
Coxsackie
Antiviral
Medium/medium
Good
Poliomyelitis
Poliovirus
None
0/Low
Stable
Measles
Paramyxovirus
Supportive
Medium
Good
Herpes
Herpes simplex
Aciclovir
High
Good
Rabies
Lyssavirus
Vaccine
Low/medium
Fatal
HIV/AIDS
HIV-1/2
Heart
Medium/high
Good to poor
Fungus
Cryptococcal meningitis
Cryptococcus
Amphoterecin B
Low
Poor to good
Toxin
Botulism
Clostridium B
Neutralizing
Low
Good if treated in time
Tetanus
Clostridium tetani
Antibodies
Low/high
Virus
sc
Prion
Mad cow disease Creutzfeldt-Jacob Kuru
PsP prion
None
Low Low Low
Universally fatal
Protozoa
Malaria Primary amoebic meningoencephalitis
Plasmodium Naegleria fowleri
Artemisinins Amphoterecin B
Low/high Low
Good Fatal
Metazoa
Neurocysticosis
Taenia solium
Praziquantel
Low/medium
Good
Prevalence: low ≤ 10,000; medium = 10,000–100,000; high ≥ 100,000.
one in 100 children develop meningococcal meningitis, the most deadly form of meningitis that has been largely eradicated in the United States. The telltale signs of bacterial meningitis include high fever (101–103 °F), sudden-onset headache, and a stiff neck, possibly associated with nausea and followed by seizures and decreased consciousness, even coma. These infections are life-threatening, can cause long-term neurological deficits, and require immediate recognition and treatment. Bacteria typically enter the brain via the circulating blood, where they are protected from lysis by neutrophils through a polysaccharide capsule. They enter into the cerebrospinal fluid (CSF) via the choroid plexus epithelium, a structure at the top of the third and fourth ventricles involved in the production of the CSF (Box 1 in Chapter 2) that bathes the brain. Once in the CSF, bacteria rapidly colonize in a compartment that is typically free of bona fide immune cells. Once associated with the meningeal covering of the brain that delineates the arachnoid space, the lysis of the bacterial wall and release of cell wall components such as lipopolysaccharide trigger a relatively nonspecific inflammatory response. This involves microglial cells and invading T-cells and is associated with the release of proinflammatory cytokines and
chemokines, including tumor necrosis factor (TNF)-α and interleukin (IL)-1β. These factors increase the permeability of vascular vessels, causing vascular edema and entry of blood-borne immune cells into the subarachnoid space. This amplifies the immune response, starting a vicious cycle. Worsening edema causes intracranial pressure to increase, inducing severe headaches, alterations in the level of consciousness, and, eventually, coma. Unequivocal diagnosis of bacterial meningitis requires bacterial cultures from CSF obtained by spinal tap. This, in addition to ruling out other causes such as fungal or viral infections also aids in the selection of the most appropriate antibiotic. However, treatment cannot wait this long and must begin immediately using empirical antibiotic treatment before CSF findings are known because mortality ranges from 3% to 20%, even with aggressive intravenous antibiotic treatment. In the developed world, viral infections of the meninges are 10 times more common than bacterial infections, with approximately 2.8 Mio cases/year globally and 75,000 cases/year in the United States; they show identical symptoms, but CSF samples contain an abundance of white blood cells and, obviously, an absence of bacteria. The vast majority of cases (85%) are caused by enteroviruses such as Coxsackie. Less common are West Nile
III. SECONDARY PROGRESSIVE NEURODEGENERATIVE DISEASES
3 Clinical Presentation/Diagnosis/Epidemiology/Disease Mechanism
virus, arboviruses transmitted through mosquitos, HIV, and, rarely, neurotropic herpes viruses such as herpes simplex virus (HSV). Because antibiotics are ineffective for viruses, treatment is palliative, and the inflammation typically resolves spontaneously. Prognosis is usually good for most cases of enteroviral meningoencephalitis, and recovery is almost always complete. In cases of more severe involvement of the brain parenchyma, the clinical outcome can be complicated by long-term neurological deficits. In some 20,000 patients annually, the viral infection is not restricted to the meningeal coverings but affects the underlying brain as well, causing inflammation that results in diffuse neurological symptoms ranging from mild cognitive impairment to severe neurological conditions such as seizures and ataxia. The resulting encephalitis can be caused by hundreds of different viruses whose identity often remains obscure. Some cases caused by HSV can be effectively treated with antiviral drugs such as acyclovir, whereas others such as poliomyelitis cannot be treated at all. In addition to bacteria and viruses, fungi can also be the cause of meningitis. Although rare among the general population, fungal infections represent the second most common opportunistic infection in HIV-infected individuals in the developing world and represent important infections in transplant patients and other individuals with weakened immune systems. Unlike bacterial and viral meningitis, which present with acute severe headaches, fungal infection of the meninges presents with slowly developing subacute symptoms that worsen over the course of several weeks. Mortality is high (30–70%), and treatment with an antifungal agent, such as amphotericin B, is often ineffective if the infection is of long duration.
3.2 Botulism Botulism is an illness caused by a bacterial neurotoxin rather than a bacterium itself. It typically results from eating spoiled food that contains the bacterial toxin. After the gastrointestinal symptoms of food poisoning subside, a characteristic flaccid paralysis (loss of muscle tone) develops. Botulism toxin is produced by gram-negative bacteria of the genus Clostridium, which are abundant in the soil or in silt of streams and lakes. Bacterial spores are inert and survive for many years. When conditions are favorable, spores germinate and bacteria grow. Warm, moist temperatures and anaerobic, oxygen-free conditions activate the bacteria to produce toxins. Such conditions occur, for example, in homecanned food items, sauces, ketchup, and the like. There are seven different botulism toxins, designated A–G, each produced by a different strain of Clostridium. These toxins are the most toxic substances known to man; an amount fitting on the tip of a needle is sufficient
239
to kill an adult, and half a pound is sufficient to kill every human on earth. Only A, B, E, and F toxins cause disease in humans. Different geographic locations correlate with different toxin outbreaks; type A is the most prevalent in the western United States (California, Washington, Colorado, and Oregon), type B is most common in the eastern United States, and type E is most common in Alaska. This distribution correlates with the most prevalent Clostridium bacteria or their spores found in soil samples from these geographic regions. Spores are inert and can survive for years. In general, we assume that botulism is the result of dietary intake of the toxin. However, in addition to foodborne botulism, there are two additional distinct clinical presentations of botulism. Wound botulism is caused by Clostridium colonizing a contaminated wound, where the toxin enters the bloodstream at the site of injury. Furthermore, children and adults can develop botulism as a result of Clostridium colonization of the gut, where the assiduous release of toxin causes paralytic symptoms in the complete absence of gastrointestinal signs. Indeed, in the United States, childhood botulism is the most common form of the disease. The source of contamination is not entirely clear, but in 20% of affected children, exposure to the spores could be traced to honey that was applied to pacifiers. However, soil, dust, or other contaminants are likely the predominant source of the bacterial spores. Infantile botulism may contribute to some of the unexplained causes of sudden infant death syndrome. The botulism toxin originates as a large, 150-kDa precursor polypeptide, which, upon bacterial lysis, is activated by proteases. The active toxin consists of a 100kDa heavy chain and a 50-kDa light chain linked by a disulfide bond (Figure 2). The toxin enters cholinergic nerve terminals via binding to two receptors: a ganglioside and, depending on the toxin type, either synaptotagmin or the synaptic vesicle protein SV2.2,3 The latter are typically inside synaptic vesicles. Upon fusion with the membrane to release transmitter, the light chain of the toxin binds to either synaptotagmin or SV2, which now transiently appears on the cell surface (see Figure 3). Upon vesicle recycling, the entire toxin becomes trapped in the vesicle. Because of the acidic environment, a conformational change of the toxin allows its heavy chain to integrate into the vesicle membrane, where it forms a pore. This pore is used for the light chain to leave the vesicle and enter the cytoplasm. Here it binds to and cleaves the vesicle-associated membrane protein (VAMP) and synaptosomal-associated protein (SNAP25). These molecules belong to a protein complex called SNARE that mediates Ca2 +-dependent vesicle docking and fusion. The B, D, F, and G toxins exclusively cleave VAMP, whereas the A, C, and E toxins cleave SNAP25 (Figure 4). The end result is the same, namely,
III. SECONDARY PROGRESSIVE NEURODEGENERATIVE DISEASES
240
10. Infectious Diseases of the Nervous System
FIGURE 2 Botox/B binding to both synaptotagmin and the ganglioside receptor on the membrane of cholinergic neuron. With permission from Ref. 2.
i nsufficient acetylcholine release, causing muscle paralysis. Note that the rate of toxin uptake is a direct function of the synaptic activity as vesicle fusion is an essential step in toxin uptake. As a consequence, more active muscle groups paralyze faster. Tetanus toxin (discussed below) is produced by a related strain of Clostridium and is taken up in a similar fashion but transported into the spinal cord, where it disrupts the release of γ-aminobutyric acid (GABA) from interneurons, causing spastic as opposed to flaccid paralysis. The clinical presentation of botulism, regardless of the source, is rather similar. In food-borne botulism, initial symptoms occur within 18–36 h after exposure to the toxin, yet incubation can range between 6 h and 10 days. As the toxins cause sustained blockage of acetylcholine release, they affect both muscarinic and nicotinic cholinergic neurons. Hence, in addition to the nicotinic skeletal, facial, and respiratory muscles, the muscarinic autonomic and smooth muscles are also paralyzed. Facial and throat muscles are the first skeletal muscles affected, and paralysis progressively descends to include the trunk and the respiratory and visceral muscles. Dry mouth, blurred vision, and diplopia (double vision) are also early symptoms, and dilated fixed pupils are typical. This is followed by drooping eyelids (ptosis), a hoarse voice, overall difficulty speaking (dysarthria), and difficulty understanding language (dysphasia). Severe cases affect the respiratory muscles and diaphragm, causing respiratory failure.
Food-borne botulism is rare, with about 10 cases each year in the United States. Infant botulism occurs in 80– 100 cases annually, typically in children younger than 6 months of age. It is believed that immature intestinal microflora and low acidic bile content, which normally suppress Clostridium growth, increase susceptibility. As the toxin enters the blood stream, these infants display symptoms including apathy, weakened cry, loss of appetite, constipation, and overall weakness (floppy baby syndrome), which develop over days to weeks. Less than 1% of infants die from botulism, and survivors have no long-term sequela. Treatment typically involves intravenous injection of an equine antitoxin containing either antibodies to A, B, E, or A–G toxins and that neutralizes molecules not yet bound to receptors. A single injection of 10 mL of antitoxin is sufficient to stabilize the disease. Recovery occurs over time through the regeneration of nerve terminals; it typically takes 2–8 weeks. The cure rate is high, and most individuals are left without any lasting deficits. It is important that suspected cases be reported to the US Centers for Disease Control and Prevention to ensure that potential sources of the toxin can be identified and destroyed. Poisoning by food-borne toxins can be prevented by boiling; however, this requires a temperature of 121 °C, for example, in a pressure cooker for > 3 min. It is important to stress that botulism is a rare neurological condition. Not counting infant botulism, fewer than
III. SECONDARY PROGRESSIVE NEURODEGENERATIVE DISEASES
3 Clinical Presentation/Diagnosis/Epidemiology/Disease Mechanism
241
FIGURE 3 Model for the entry of Botulinum Neurotoxin (BoNT) into nerve cells. SNARE proteins mediate vesicular docking. Vesicular Synaptobrevin (Syb) and membrane Syntaxin (Syn) and SNAP-25 must interact (A) for vesicular docking on the cell membrane, which in turn allows vesicular release of acetylcholine (ACh) (A). Once released, ACh binds acetylcholine receptors (AChR), which mediates muscle contraction (C). The intravesicular parts of synaptic vesicle proteins become exposed on the plasma membrane as the vesicle integrates into the membrane (D). When BoNT is present, it binds to a ganglioside molecule on the cell surface, which allows BoNT to access their protein receptor (exemplified here for synaptotagmin (Syt)) (E). After binding the protein receptor, the neurotoxins are endocytosed via retrieval of synaptic vesicles (F). The lumen of the recycling vesicles becomes acidified via the action of the vesicular proton pump (G). Acidification provokes a structural rearrangement in the neurotoxins, whereby the heavy-domain (HC and HN) forms a channel through the vesicular membrane (H). The light chain (L) passes through the channel due to partial unfolding and is released to the cytosol following reduction of the disulfide bond (I). Ultimately, the light chain cleaves its target SNARE(s), Syb, Syx, or SNAP-25, (J) and thus blocks the synaptic vesicle cycle, which it exploits for cellular entry. Preventing the further docking of ACh vesicles prevents ACh release and subsequent binding to AChR, leading to paralysis (K). The entry of BoNT into nerve cells requires the vesicular docking cycle to expose the appropriate protein receptor (here shown as Syt); hence the rate of toxin uptake is directly related to vesicular fusion and therefore directly related to synaptic activity. This results in more active muscle groups becoming paralyzed more quickly. Courtesy of Stephanie Robert and adopted from Ref. 4.
2500 cases have been reported in the United States since 1899, and the average number of disease outbreaks is around 10 per year, each affecting (on average) 2.6 people. Although historically botulism has been a deadly disease, with 60% of infected individuals dying, the introduction of antitoxins and improved supportive care, including mechanical ventilation, has reduced case fatality ratios to less than 1%.
3.3 Tetanus Tetanus toxin5 is similar to wound botulism in that it is a bacterial toxin produced by bacteria that germinate from bacterial spores found abundantly in the
environment and that enter the body through open wounds or scrapes. Clostridium tetani produces two toxins, tetanolysin and tetanospasmin. Because only the latter causes the typical symptoms, it is the one simply referred to as tetanus toxin. As with botulinum toxin, tetanus toxin is produced as a 150-kD precursor protein with a light (50 kD) and heavy chain (100 kD) linked via disulfide bonds. The heavy chain binds to gangliosides on presynaptic motor terminals and mediates the entry of the complex into the cell. From there, it is transported (retrogradely) back to the motor neuron cell body via axonal transport, reaching the motor nuclei in the ventral horn of the spinal cord or motor nuclei of the cranial nerves in the brainstem. The major d ifference
III. SECONDARY PROGRESSIVE NEURODEGENERATIVE DISEASES
242
10. Infectious Diseases of the Nervous System
FIGURE 4 The various botulinum toxins have specific presysnaptic targets, which are all involved in synaptic vesicle release.
between botulinum toxin and tetanus toxin is that the former stays at the synapse and inhibits transmitter release, whereas the latter is transported back to the neuronal cell body and ultimately enters “one cell up” in the presynaptic terminal. How the toxin escapes lysosomal degradation and instead leaves the cell to enter its presynaptic GABAergic terminal is not well known, but transcytosis is a suspected mechanism. Once in the GABAergic terminal the light chain of the toxin, which is an endopeptidase, cleaves VAMP2 (synaptobrevin), thereby preventing the release of the inhibitory neurotransmitter GABA. This results in unregulated excitatory activity of the motor neurons, causing the tetanic muscle contractions that are the hallmark of the disease. Hence the difference of the site of action between these toxins makes one induce flaccid paralysis and the other spastic paralysis. Depending on the motor nuclei affected, tetanus toxin infection can present with a wide variety of symptoms. It can cause spastic contractions of just about any muscle group. If motor nuclei innervating the laryngeal or respiratory muscles are affected, the infection is life- threatening because respiratory failure is likely; indeed, this is the most common form of death resulting from tetanus poisoning. Although deaths related to tetanus are exceedingly rare in the Western world, where vaccination is common, it is estimated that tetanus kills about 100,000 people worldwide annually. Increased global access to
accination has reduced total tetanus-related deaths v from ~ 1 Mio in 1980 to fewer than 100,000 in 2020. Improved antiseptic conditions during childbirth have reduced the number of neonatal deaths due to tetanus from 787,000 in 1988 to 30,800 in 2017. Once symptomatic, wound cleaning and administration of antitoxin, where available, along with benzodiazepines to control spasms, can stabilize the disease. A ventilator may become necessary if breathing problems develop. If caught in time, the prognosis is typically good, with a complete reversal of symptoms.
3.4 Neurosyphilis Neurosyphilis (Tabes dorsalis)6 is a late presentation of syphilis that occurs in 10% of infected and untreated individuals. Syphilis derives from the Greek word syphlos, meaning crippled or maimed, since untreated patients can develop bizarre-looking cutaneous, noncancerous growths on any parts of their bodies, called gummas (Figure 5). Syphilis is a bacterial infection that is typically acquired through sexual contact and hence belongs to the sexually transmitted disease category. The infectious reagent is Treponema pallidum, a gram-negative bacterium called a spirochete because it looks like a corkscrew. The infection typically occurs through contact with open wounds or fluids during unprotected sex. Since the disease can be effectively controlled using the antibiotic
III. SECONDARY PROGRESSIVE NEURODEGENERATIVE DISEASES
3 Clinical Presentation/Diagnosis/Epidemiology/Disease Mechanism
FIGURE 5 Gummas disfiguring the head and face of an individual with syphilis. Bust in Musée de l’Homme Paris.
penicillin G, incidence rates have decreased significantly since the 1940s in the developed world. Infection rates are at epidemic levels in the developing world with 6 Mio new cases reported globally each year. Infection rates are also on the rise in the developed world presumably because of unprotected sex among men infected with HIV. The disease usually involves three phases. The primary infection follows the inoculation of an individual with about 500–1000 bacteria. Within 36 h, these replicate and result in a painless ulceration called a chancre. These typically occur in the genital areas. After 2–6 weeks the second stage of the disease continues, with wide infiltration throughout the body and nervous system. Afterward, during the latent stage of disease, patients are frequently asymptomatic for many years. About 10% of patients with untreated syphilis develop neurological symptoms called neurosphyilis, or tabes dorsalis, 10– 15 years later. These include progressive muscle weakness, unsteady gait, and cognitive impairments typically including confusion, delusions, disorientation, and depression. Pupillary abnormalities (Argyll-Robinson pupils), dysarthria, and tremors in extremities can also be present. The earliest stages of neurosyphilis involve inflammation of the meninges presenting with headache, nausea, vomiting, and, occasionally, seizures. This acute
243
s yphilitic meningitis responds well to aggressive penicillin treatment, whereupon the symptoms typically resolve completely. The classical neurosyphilis presentation was described by Romberg and includes the following “3 Ps”: paresthesia, pain, and polyuria. Patients experience sharp, lightning pains (paresthesia) and have an unsteady ataxic gait. With eyes closed, patients typically cannot walk or even stand upright, but fall. This is called a Romberg sign and is caused by a loss of proprioception. Unlike the senses with which we perceive the environment, feedback information from stretch receptors in skeletal muscles tells us the relative position of each extremity in space. Such receptors also report the state of contraction of visceral muscles such as the bladder, explaining polyuria (enhanced frequency of voiding). The progressive degeneration of the spinal sensory nerve roots along the dorsal column of the spinal cord essentially causes a functional deafferentation, as seen in Figure 6. This includes the sensory nerve fibers innervating the skin and the viscera, causing pain comparable to amputation or diabetic neuropathies. Such deafferentation typically causes hyperexcitability in the ascending pain pathways and is irreversible. A number of studies suggest that approximately onethird of individuals infected with syphilis show bacteria in their central nervous system (CNS) and are at risk of developing neurosyphilis. In contrast to the previously held assumption that bacterial spread to the CNS occurs late during the disease, this is not supported by research, which instead suggests that CNS infection occurs early in disease, possibly within just a few days after infection. The natural history of neurosyphilis was comprehensively evaluated in the infamous Tuskegee studies that enrolled a large number of infected individuals in rural Alabama. This clinical study, conducted from 1932 to 1972 by the US Public Health Service, enrolled 600 sharecroppers, of whom 400 were infected and 200 were not. In spite of the introduction of a curative agent midway
FIGURE 6 Cross-section of a spinal cord showing loss of dorsal spinal neurons in the dorsal spinothalamic tracts (white areas encircled by dashed line) in a patient with neurosyphilis. Centers for Disease Control and Prevention/Susan Lindsley.
III. SECONDARY PROGRESSIVE NEURODEGENERATIVE DISEASES
244
10. Infectious Diseases of the Nervous System
through the study in 1945, disease-carrying patients were never offered treatment, and the disease was allowed to run its course. This egregious lapse in ethical conduct led to the Belmont report, which in 1979 established the procedural and ethical framework for human experimentation, including the requirement for informed consent and oversight by an independent review board. Since 1945, syphilis has been effectively treated with penicillin, and the bacterium has not developed any resistance over the past 70 years. High doses are necessary to reach the CNS and therefore penicillin G is given through daily infusions over a 10–14 day period. Two serological tests are available (Rapid Plasma Reagin and Venereal Disease Research Laboratory) that can detect established infection with high accuracy. CSF is analyzed only if neurosyphilis is suspected. Such infection presents with an increased number of white blood cells (> 5 per ml) and CSF protein, which is normally very low and increases to > 45 mg/dL. Additional serologic tests for syphilis should show reactivity.
3.5 Poliomyelitis Poliomyelitis, also known as infantile paralysis, is caused by infection with the poliovirus, a nonenveloped RNA enterovirus that colonizes the gut. The virus is contracted through the mouth and, as an RNA virus, highjacks the body’s own cells as it colonizes, coaxing cells into producing more virus. The virus is highly resistant to acidic environments and can survive for long periods of time in sewage or water. Hand-to-mouth contact is the typical mode of spread. Once ingested the virus enters the bloodstream, and 99% of infected people show either no response or only a short, mild fever. In 1% of patients, however, the virus enters the brain and causes an inflammatory response. The virus infects and kills motor neurons in the spinal cord, motor cortex, and brainstem. This typically occurs long after the virus has colonized the body. The virus shows an intrinsic neurotropism and specifically selects motor neurons over sensory neurons and neurons over glia cells. The entry of the virus into the nervous system can occur via two routes: by crossing the blood–brain barrier or infecting the peripheral nerves, which transport the virus back to the spinal cord and brainstem via axonal transport. CD155, also known as the poliovirus receptor, can bind to a protein that is associated with the dynein complex, which moves cargo along the axonal microtubules. Once the virus has entered the brain though this “Trojan horse” strategy, it is free to replicate and infect additional neurons. Ultimately, the virus kills its host neurons by shutting down the cell’s protein biosynthesis and activating the apoptotic caspase cascade. Remarkably, the virus primarily affects the motor pathways responsible for the lower extremities and completely spares the sensory neurons.
Most polio cases occur in infants, rarely babies or adults. Until the mid-1950s, polio crippled on average 40,000–50,000 children each year in the U.S. Through aggressive vaccination, polio has been essentially eradicated in the Western world, but it is still found in pockets of the developing world, particularly in Afghanistan, Pakistan, and West Africa. There are three poliovirus serotypes, all of which can cause disease although type 1 is the one most often associated with paralytic disease. All three serotypes are contained in modern vaccines. Global vaccination campaigns have eradicated type 2 polio in 2015 and type 3 in 2019 with hopes of complete eradication in the near future. There is no cure, and treatment is entirely supportive. Death is rare, but permanent disability affecting primarily the lower extremities is the norm (Figure 7). In some patients who have had stable disease for 30–40 years, muscle weakness can suddenly worsen, causing a poorly understood postpolio syndrome. There is no evidence of residual virus at this stage, and in all likelihood the remaining motor neurons are simply exhausted from overuse.
3.6 Rabies A much more common RNA virus is the rabies virus,7 which typically infects animals but can spread to humans through animal bites. Frequent carriers include bats, raccoons, skunks, and foxes, and disease is typically transmitted to people via stray dogs. Rabies is found throughout the world except Antarctica. After entering the body via a bite, the virus replicates in the surrounding muscle tissue. From there, the virus spreads centrally along peripheral nerve fibers, presumably via fast axonal transport. After the initial peripheral to central spread, the virus eventually replicates in the acinar cells of the salivary gland, which become the vector for
FIGURE 7 Children with polio at the Amar Jyoti Research Center, Delhi, India. Photo: WHO/P. Virot (UN News Center).
III. SECONDARY PROGRESSIVE NEURODEGENERATIVE DISEASES
3 Clinical Presentation/Diagnosis/Epidemiology/Disease Mechanism
disease transmission. Rabid animals have excessive saliva containing the virus, often seen as froth. As a result of this excessive salivation, infected animals or people have difficulty swallowing and show hygrophobia (fear of water). The CNS inflammation is mild and there is surprisingly little evidence of neuronal cell death given the severity of the disease. On pathological specimens, inclusion bodies, called Negri bodies, can be found in cerebellar Purkinje cells and pyramidal neurons. It must therefore be assumed that functional rather than anatomic changes explain the neurological symptoms. Rabies-infected animals and humans become aggressive and fearless and are quite dangerous to uninfected animals or humans, yet the biological underpinning of the “rabid” behavior is not understood. The disease is typically first noticed by an infection around bite marks, although bat bites, common for transmission in the United States, may be too small to be identified, making a suspected diagnosis difficult. In 80% of patients the disease is encephalitic and involves the forebrain and cerebellum, causing furious behavior with combativeness and hallucinations. In 20% of cases, it is paralytic, causing flaccid paralysis that is easily misdiagnosed as Guillain-Barré syndrome. Rabies is almost always fatal; patients die within days of becoming symptomatic. However, if recognized early after infection during the incubation period, which lasts from 1 to 3 months, proper wound care and postinfection vaccination may be able to contain the disease. Given the extensive vaccination of animals and particularly pets in the United States, rabies rarely infects humans. However, canine-transmitted rabies is common in Asia and Africa, where an estimated 55,000 people die from it each year.
245
3.7 HIV and Neuro-AIDS HIV is one of the most common viral infections worldwide and the single most important virus-induced neurological disease thus far described.8 Globally, approximately 1.7 new cases of HIV/AIDS are reported each year, with an estimated 40 Mio people living with HIV/AIDS in 2020. HIV is a member of the retroviral family, which is characterized by a unique replication cycle whereby genes are encoded by RNA, rather than DNA, and must undergo reverse transcription and integration into the host DNA. Retroviruses almost exclusively infect vertebrates. Because they insert their genome into the host cell DNA, they are adept at regulating a host cell’s behavior. The viral core contains two copies of single-stranded RNA and is surrounded by a capsid and enveloped by a lipid-rich membrane (Figure 8). The viral RNA is reverse transcribed into double-stranded DNA in the cytoplasm and trafficked to the nucleus, where it is irreversibly inserted into the host cell’s genome. Some of these genes initiate the production of virions by the host cell, which then are released from the plasma membrane by budding (Figure 8). The host cell then has been reprogrammed to produce viral genes and is entirely under the regulation of the virus. HIV has been unequivocally determined to be the cause of acquired immune deficiency syndrome (AIDS). Although there are two distinct strains of HIV, HIV-1 and HIV-2, which differ genetically, the vast majority of AIDS cases are caused by HIV-1. AIDS is a relatively new disease, dating back to late 1970s, when the virus must have jumped species from chimpanzees or gorillas to humans. The virus was first isolated in 1985 and is probably the most well-studied pathogen. HIV typically infects T-lymphocytes by attaching via its gp120 coat protein to the CD4 receptor. This results in a
FIGURE 8 Human immunodeficiency virus I budding from CD4 lymphocytes (left). Schematic of the structure of the viral particle (right). Reproduced with permission from Ref. 8.
III. SECONDARY PROGRESSIVE NEURODEGENERATIVE DISEASES
246
10. Infectious Diseases of the Nervous System
conformational change, allowing the complex to bind two coreceptors, CXCR4 and CCR5, leading to internalization through fusion with the cell membrane. Once inside the T cell the virus releases its RNA, which is transcribed into DNA and integrates into the host genome. The reprogramming of the T cell causes production of virions that are released from the lymphocytes by budding. HIV is primarily transmitted through sexual contact, and the likelihood of infection is directly proportional to the viral load of the infected individual. Transmission through infected blood products is also possible but is, in light of recent extensive testing, exceedingly rare (one in one million). However, the risk of health care workers contracting HIV from job-related injuries is significant. Close to one million accidental needle sticks occur each year in the United States, each carrying a 0.3% chance of infection if the needle was used by an HIV-infected individual. There is a 20–30% likelihood that a pregnant HIV-infected woman will infect her baby. The global scope of HIV is immense, approaching a current incidence of 40 million people. Almost two-thirds of infected individuals are in sub-Saharan Africa, where the infected are not predominantly homosexual men, as is the case in the United States. Instead, 50% of all HIV cases worldwide are women, and 2.5 million children are HIV positive. Of the 1.1 million people confirmed to be HIV positive in the United States, 50% are homosexual men and 10,000 are children. The systemic disease presents with profound immunodeficiency secondary to uncontrolled production of the virus and loss of T-lymphocytes. Over a period of a few weeks, fever, headache, swollen throat, and lymph nodes, as well as severe muscle pain and fatigue, develop and last for several weeks. This acute stage of disease is followed by a latent stage lasting from a few months to well over 20 years. In the final stage of the disease, most of the body’s T cells have been lost, resulting in a drastic reduction in the blood concentration of T-lymphocytes ( 42 °C
Heat, chili-capsaicin, garlic
Anandamide, ammonia Ca2 +, Mg2 +, LPA, NO
TRPV2
> 52 °C
Cannabidiol
IGF, lysophosphadylcholine
TRPV3
25–60 °C
Camphor, menthol
TRPA1 (Ankyrin)
1000 patients and compare the drug with other treatments that are considered the current gold standard for a given condition. In a multiple sclerosis trial, for example, this could be a comparison to interferon-β. The end point of the phase III study is likely different from the phase II study and seeks a significantly improved outcome over other available therapies. Phase III studies may also consider combined use of a new drug with an existing drug. Successful completion of phase III and drug approval by the FDA is required for a drug to go to market. Approval is issued only for the particular disease indication for which the drug was studied. In the following years of widespread clinical use, some drugs undergo a phase IV study that examines unexpected drug interactions or looks at the response of patient population to the drug. For example, an epilepsy drug may be monitored for interaction with beta- blockers used in patients with heart disease.
4.3 Important Considerations The entire plan of a study, including its objective, enrollment, primary and secondary end points, methods of collecting, handling, and analyzing data, and expected sample size needed at each stage, is defined up front and forms the clinical trial protocol. This protocol must be approved by an institutional review board (IRB) and the FDA. Enrollment Unlike in animal studies, where most animals are the same except for sex and age, a human study has to thoroughly consider the appropriate target population that could and should benefit from the treatment, as well as a well-matched comparison group. This is not always trivial, and use of the wrong comparator or lack of a control group may easily send a trial off course. Moreover, a balanced approach must be taken to ensure inclusion of patients of all races, both sexes, various ages, and other variables, unless a strong argument can be made for the exclusion of a certain subgroup. Typically, early clinical trials exclude children because they cannot give informed consent, unless the condition of interest is exclusively a childhood disease. End Points Another important consideration is meaningful end points to be studied. In animal experiments for an antiseizure medication, for example, one may measure seizure frequency determined by continuous electroencephalography recordings. This is readily feasible in
VII. BENCH-TO-BEDSIDE TRANSLATION
4 What Are Clinical Trials and Why Do Them?
laboratory animals but not in humans since we cannot expect people to be in the hospital for weeks to receive continuous monitoring. Hence we may select a self-reported seizure log as a substitute. In addition, a secondary end point may be defined. This could be cognitive improvement, which would require a battery of behavioral cognitive tests. This greatly complicates the study and introduces additional confounding factors. Sample Size Determining the correct sample size before beginning the study is an important component of a successful trial and is the most commonly flawed aspect of trials that are reviewed throughout this book. We often talk about a study being underpowered to show significance for an expected end point. Before each phase of the trial, the necessary population must be calculated with the help of a statistician through empirical power size calculation rather than an open-ended experimentation. The size of the study population depends on the expected effect size, with a larger effect requiring a smaller population for study. It makes intuitive sense that detecting a 5% decrease in average infarct volume after a stroke intervention is more difficult to detect than an increase in the number of patients surviving for at least 12 months, and thus would require a larger population for study. Sample size is also affected by the effect size of the control or placebo group and the expected dropout rate in the trial. Excellent example calculations are found in the literature.4 Patient Safety and Informed Consent Human experimentation has, sadly, not always been conducted ethically. The violations are many, ranging from the brutal atrocities performed by Nazi doctors during World War II to the well-intended experimentation with the polio vaccine by Jonas Salk on intellectually disabled children living in orphanages without their consent. One objectionable study that has become a mandatory read for the scholar of clinical trials is the Tuskegee Syphilis Study, which ran from 1932 to 1972 and was sponsored by the US Department of Health. It examined the effects of untreated syphilis in 400 African American men. Researchers withheld treatment even when curative penicillin G became widely available. Most importantly, the study participants were never told that they were participating in an experiment.5 Several important safeguards have now been implemented to ensure the safety of those participating in clinical studies. Foremost is the requirement for voluntary participation and informed consent and the assurance that a participant can withdraw at any time for any reason or for no reason whatsoever. Moreover, an IRB reviews each trial well ahead of its approval. These boards
423
typically include lay people from the community who scrutinize the objective of a trial. The IRB ensures that the expected benefits far outweigh the risks to participating individuals and that the study has clear conditions under which a trial will be ended. The IRB also ensures that participants are well informed and consent to the study. IRBs have become a major administrative hurdle to clinical trials and are perceived by some investigators as overreaching. However, given the history of human experimentation, it is important to err on the side of patient benefit and safety. Reporting and Registration In the US all clinical trials must be registered with clinicaltrials.gov and all pertinent information publically displayed. The investigator also must ensure that the outcome of the data are published. Unfortunately, this requirement does not require publication and wide dissemination in a peer-reviewed journal, only an abbreviated summary on this Website. In my own experience, these summaries are often difficult to interpret. Success and Cost As of October 2020, there were approximately 150,000 drug trials registered. Note that these are not 150,000 unique drugs being evaluated. Instead, many trials are testing the same drug for different disease conditions, or combination with other drugs. As already alluded to, few of these drugs are truly novel compounds. Many are recycled for use in a different disease indication. The entire process of moving drugs from discovery to market, along with success, failure, and costs, is illustrated in Figure 1. This graph provides startling news. For every 10,000 drugs that enter the discovery phase, only one drug makes it to market, and most drugs that begin preclinical testing fail. For those that remain, the cost to market is enormous. As of 2020, the cost to develop a single drug was $2.6 billion. The whole process takes 15 years on average, with the clinical testing phase lasting 6–8 years. Consider that current patent protection in the US is typically only 17 years from the time of issuance. Hence drug companies have only a few years to market their drug with exclusive rights before generic drugs are admitted on the market. Moreover, a drug company must fund all its failed drugs through profits on the few successful drugs. This explains the high cost of most newly developed drugs; drug companies must recoup the cost of drug development in the few years during which they can exclusively market the drug. Also, much of the development cost is the same whether it is for a large or a small patient market, making it difficult to justify developing treatments for rare diseases with small markets.
VII. BENCH-TO-BEDSIDE TRANSLATION
424
17. Drug Discovery and Personalized Medicine
FIGURE 1 The timeline of a clinical trial along with the relative success rate, number of patients involved in each phase, and overall cost estimates. Mio = million.
5 THE PLACEBO EFFECT As already mentioned, a clinical trial should ideally use a placebo control group rather than a nontreatment group if possible. A placebo is typically an inert substance given to a patient in lieu of an actual medication. While it is often thought of as a “sugar pill,” a placebo may be better defined as an intervention designed to simulate medical therapy, but it is not considered to be an actual treatment at the time of its use. This definition broadens a placebo to the use of devices and even sham surgery in treatment. The use of placebos in medicine was quite common throughout the 18th and 19th centuries. Indeed, in 1807, President Thomas Jefferson reported that one of the most successful physicians at the time confided that he “uses more bread pills, drops of colored water and powders of hickory ashes than all of other medicines together.”6 Until the 1950s, a placebo was considered a “humble humbug” that comforted a patient without doing any harm. Then, however, it became elevated through a study published by Beecher7 in 1955 showing an overall 35% effectiveness of a placebo compared to an actual treatment. This, along with reports that sham surgery can provide medical relief, defined what we now call the placebo effect, namely clinical improvements achieved in the absence of an active intervention. Of course, in hindsight, many of the potions provided by traveling doctors and self-anointed healers throughout the centuries provided little more than a placebo effect. Today, the placebo has become the most important control in defining whether a drug or intervention is actually curative per se or if the improvements are the result of the environment and belief system in which the treatment occurs. The placebo effect must not be dismissed as humbug. It is real, it is strong, and it is particularly applicable to neurological illnesses. Our understanding of the neurobiology underlying the placebo effect is incomplete, but it is clear that our mind strongly influences
brain and body function through the release of neurotransmitters and hormones. The placebo effect has a strong psychological component and is therefore probably only effective when the patient is convinced that (s)he receives an active treatment as opposed to a placebo. Interestingly, however, patients may improve even when told that they are receiving a placebo, but whether these patients actually trusted this information or persisted in the belief that they were receiving an active treatment is unknown. While placebo treatment works for many ailments, it does differ in efficacy and is strongest for conditions with a large psychological component, such as pain, anxiety, and depression. Functional neuroimaging studies done in conjunction with the use of placebo analgesics have been particularly insightful. These studies demonstrate that the belief that an analgesic has been applied before inflicting a painful stimulus is sufficient to suppress the activation of nociceptive pathways in the spinal cord.8 This effect is most likely mediated through the release of endogenous opioids in the spinal cord in response to cortical activation of the descending pain control systems in the brain stem; indeed, opioid antagonists block the placebo effect. These experiments show that the placebo effect is mediated by an actual physiological change that is identical to that of an active drug, a truly remarkable observation. One of the most pronounced placebo effects is observed in the study of antidepressant drugs. When antidepressants drugs such as Prozac were compared to placebo, they were found to be equally effective at improving mood. Importantly, just as with placebo analgesics, for patients who responded to the placebo, it affected the very same regions in the limbic brain involved in experiencing sadness as did the antidepressant drug, causing an identical change in glucose utilization imaged through positron emission tomography.9 This once again shows a physiological effect of the placebo that is identical to that of the active drug.
VII. BENCH-TO-BEDSIDE TRANSLATION
6 Why Do Clinical Trials Fail?
These placebo studies present us with quite a dilemma. Depressed patients who are untreated stay depressed. However, 75% of those who take medication get better, but so do the majority of patients who take the placebo. In fact, the difference in effect size is only 1.8 points on a 51-point behavioral scale and would thus be considered clinically meaningless. In short, placebos are as effective as antidepressants, yet antidepressants can have dramatic side effects including weight gain, fatigue, loss of libido, and suicidal thoughts. Moreover, they can cost as much as $9000/year. Yet a physician cannot prescribe a placebo to a patient under the pretense that it is a real drug. This approach can be used only with informed consent in a clinical study. Until we find a solution to this dilemma, drug companies are enjoying >$10 billion/year in sales for drugs that may be no better than sugar pills. Yet another consideration that contributes indirectly to the placebo effect in clinical trials is the fact that simply being part of a medical study and in the hands of a specialist has a positive effect on treatment outcome. The simple act of showing up at the doctor’s office and being cared for by white-coated professional with countless diplomas decorating the office walls makes people feel better. Taken together with the placebo effect, it is clear that any clinical study must correct for these nonspecific effects to ascertain a drug’s true biological activity. Unfortunately, this typically increases the number of patients required for a study, since the effect size (difference between drug and placebo) is always reduced by these nonspecific effects. For example, in the case of antipsychotic drugs given to patients with schizophrenia, this difference is a mere 17%, with placebo being almost as effective as the actual drug (24% vs. 41%).
6 WHY DO CLINICAL TRIALS FAIL? There are roughly 150,000 clinical drug studies being conducted in the United States in 2020. Unfortunately, the fast majority of them will fail before a product can make it to market. Many of these failures occur during phase I or II of the trial, where drug safety and efficacy are examined; others fail in larger, multicenter phase III studies in which larger cohorts of patients are enrolled. This enormous failure rate is unsustainable in the long run. On average, each patient enrolled in a study costs between ∼$12,000 and $20,000, and the cost of successfully moving a drug from candidate to approval is estimated at $2.6 billion. Hence drug companies are spending an exorbitant amount of money to find the few drugs that stand up to these multiple phases of clinical testing. Not surprisingly, a number of studies have examined why the failure rate is so high and have identified several flaws, most of which could be eliminated.
425
6.1 Irreproducible Laboratory Data As already discussed above, drug targets are typically identified in academic laboratories. Thousands of investigators strive to find innovate new drug targets. The competition to be first to publish is fierce and is rewarded by recognition, tenure, promotions, and additional research grants. Scientists become invested in their stories, sometimes to the point of allowing bias to creep into their work. Data points may be unjustly excluded as outliers or improper scientific methods may be used to reveal statistical significance in cases where a proper analysis would discard the finding as noise. Indeed, a recent analysis of 157 articles published in the five leading neuroscience journals (Nature, Science, Neuron, Nature Neuroscience, and Journal of Neuroscience) made the startling discovery that 50% of these articles used the wrong method for statistical analysis, and in 66% of these cases, the conclusion would not have been supported had the correct test been used.10 Therefore, it may not come as a surprise that drug companies are wary of the reproducibility of laboratory findings. As a result, they have established their own laboratories where they first attempt to reproduce the published findings and validate the targets. A recent article described that in doing so, drug companies are finding that 75%–80% of scientific findings cannot be reproduced in their laboratories.11 While they do not claim that the scientists were intentionally misreporting their findings, it nevertheless exposes a major problem that must be fixed. If we cannot trust published data, we are wasting taxpayer dollars that support such research. In 2006, an editorial by Nature stated that “Scientists understand that peer review per se provides only minimal assurance of quality and that the public conception of peer review as a stamp of authentication is far from the truth.” One problem repeatedly identified is the pressure to publish, which may result in scientists failing to spend the necessary amount of time needed to understand and validate their experimental results using multiple strategies to challenge the original hypothesis. Indeed, it has become common for scientists to “prove” their hypothesis, which, of course, as the student of science knows well, is impossible. One can only challenge a hypothesis repeatedly, and if no experiment refutes it, one gains greater confidence that it may in fact be correct. That level of confidence is often not reached in published studies.
6.2 Publication of Negative Data Unfortunately, for the same reasons explained above, publishing “negative” data, that is, those experiments that refute a hypothesis, has become difficult. They tell a “negative” story, which may happen to be the truth but carries little appeal. Journals such as Nature and
VII. BENCH-TO-BEDSIDE TRANSLATION
426
17. Drug Discovery and Personalized Medicine
Science are drawn to newsworthy findings, such as an environmental link to autism, a gene that causes schizophrenia, or new evidence that cold fusion is possible. That precludes many valid negative stories from being distributed via the most widely read journals with the highest impact. Furthermore, it carries the danger that many of these stories published in high-impact journals may in fact be wrong. Let us assume that 20 researchers work on a possible link of lead with autism. They all conduct similar studies using mouse models, and each reports their data with a significance threshold of P 65. Rare mutations in genes that process APP can cause early onset of disease, before age 65 and as early as age 35. These account for 5% of all cases. Early stages of Parkinson Disease: The early stages of Parkinson Disease are characterized by resting tremor, postural instability, bradykinesia, and rigidity. During this stage, dopamine replacement therapy is most effective. EC50: Half-effective dose of a drug or intervention. It is customary to express the effectiveness of a drug over a large dosing range to determine the maximal achievable change, the minimal dose to detect any change, and the midpoint (the dose at which 50% of the effect is achieved). Comparable useful values are LD50, the lethality dose at which 50% of test subjects (mice) die. Echolalia: Echolalia is the repeating of phrases, sounds, or sentences without any evidence of understanding their meaning, as in a parrot talking back. Edema: Swelling of tissue is referred to as edema. In the nervous system, we differentiate between cellular and vascular edema. Cellular edema is caused by water moving along with ions into cells, causing them to swell. In vascular edema, serum from the blood vessels crosses into the brain after injury or during disease and accumulates extracellularly. Electrocauterization: Electrocauterization is a method to close blood vessels through injection of an electric current into a resistive metal wire or electrode in order to stop bleeding. The metal wire or electrode will heat up as the electric current is injected; as it touches a blood vessel, the heated wire or electrode will cause the vessel’s walls to collapse and close. Electroconvulsive therapy (ECT): Also called electroshock therapy, ECT is a nonmedicinal treatment for mental illness that relies on passing electric currents through the brain, eliciting seizures with the goal to change brain chemistry. Its actual mechanism of action remains elusive, yet the approach is used with some success as a last resort in severe cases of mental illness. Electroencephalography (EEG): Electroenceph alography is a technique that allows the detection of brain activity through electrodes applied to the surface of the skull. It is primarily used to diagnose seizure disorders such as epilepsy and localize their origin. Electromyography or electromyogram (EMG): EMG is a diagnostic technique that directly measures muscle function either superficially or through insertion of a needle into the muscle. In response to electrical stimulation of the nerve, the resulting voltage change is recorded in the muscle that drives muscle contraction. The technique is called electromyography, and the resulting record is called an electromyogram.
VIII. NEUROSCIENCE JARGON
442
18. “Neuro”-Dictionary
Embolus: A “floating disaster,” an embolus is a piece of atherosclerotic plaque that has broken off the plaque and floats with the blood into finer and finer vessels. Unless degraded, it can eventually occlude a vessel, causing it to cease blood flow or, worse, burst open. Embolytic strokes are caused by such a piece of plaque in circulation. Enantiomer: Many chemicals exist in two conformations called the enantiomers -S and -R. These are mirror images of each other. While identical in terms of chemical composition, their effectiveness to bind to receptors or transporters can vary dramatically. Most often the chemical exists as a random mixture of both S and R forms. Encephalitis: Inflammation of the brain is called encephalitis. This can be caused by viral, bacterial, or fungal infections, or can be due to an autoimmune response. Endocannabinoids: Endocannabinoids are chemicals produced in our body that mimic the activity of cannabis. Like the active ingredient in cannabis, THC, endogenous cannabinoids such as anandamide bind to the cannabis receptors CB1 and CB2. Endocannabinoids are known to modulate many physiological states including mood, anxiety, pain, cognitive function, and immune response. Endopetidase: An endopeptidase is an enzyme that cleaves polypeptides or proteins at internal amino acids. In contrast, ectopeptidases cleave the terminal amino acids at either the N- or C-terminus. Enteroviruses: Enteroviruses are a diverse group of single-stranded RNA viruses that affect humans by entering the gastrointestinal tract or airways. Important examples are poliovirus, rhinovirus, and Coxsackievirus. Ependymoma: An ependymoma is a primary brain tumor arising from the ependymal cells that line the cerebral ventricles. Most commonly found in children, they can often be surgically removed and carry a favorable prognosis. Epigenetic: Epigenetics refers to persistent, but reversible, heritable changes in gene expression without alteration of the DNA itself. The gene(s) of interest are not directly affected, but the ability of the gene to be transcribed into protein is regulated via a number of modifications at the level of the chromatin or individual genes. The two most common epigenetic regulations involve the compaction of chromatin around histones, which affect whether a gene can be transcribed or not, and the methylation of individual gene promoters along the chromatin, which causes transcriptional repression. Specific enzymes interact with histone tails and loosen the chromatin by applying modification groups. For example, addition of an acetyl group (acetylation) relaxes the DNA, making it accessible for transcription. Similarly, the methyl marks on the promoters are applied by a group of enzymes called DNA methyltransferases, which typically silence the gene of interest. Epigenetic
changes can explain how environmental conditions can influence disease and behavior and confer susceptibility to disease or resistance without changing a person’s genetic makeup. Epilepsy: Epilepsy is a very common neurological disorder characterized by the spontaneous recurrence of two or more seizures. These seizures are caused by abnormal synchronous discharge of neural activity and result in a range of behavioral symptoms. Epilepsy can be the result of known or unknown inborn genetic changes, or can develop later in life as a result of a brain insult such as trauma, infections, vascular changes, or tumors. About two-thirds of patients with epilepsy can be effectively treated with a range of antiepileptic drugs, yet for the one-third of patients suffering from pharmacoresistant epilepsy, quality of life is severely affected. Epileptogenesis: This refers to a time period during which the brain acquires epilepsy. It is a prodromal phase during which an individual does not show overt signs of epilepsy; no behavioral seizures are present and EEG abnormalities may not be visible. During this time, progressive changes in the interconnectivity of neural networks lead to clusters of cells that have the potential to fire in synchrony and initiate seizures. Such seizure-prone networks must be provoked by additional intrinsic or extrinsic factors to generate a seizure, but many of these factors are not yet known. Such factors probably differ from patient to patient, but may include diet, drugs, temperature, light, sounds, smells, stress, or sleep deprivation, to name just a few. Unfortunately, little is known about this period of time preceding presentation of disease, although this is an area of active research. Episodic memory: Memories that capture episodes in one’s life in an autobiographical fashion are referred to as episodic memories. These include places, times, and emotions within the context of life stories, such as a collection of experiences that form a narrative. For example, remembering one’s wedding is an episodic memory. Epstein–Barr virus: Epstein–Barr virus is a DNA virus of the Herpes family that causes mononucleosis. It has been implicated as a risk factor for multiple sclerosis, yet it is among the most common viruses in humans, with an estimated presence in 90% of the population. Essential tremor: Often called benign tremor, essential tremor is a high-frequency, low-amplitude shaking or movements that affect hands, arms, head, or even muscles involved in articulation of speech. This is the most common movement disorder, affecting 5% of the population, with impairments ranging from mild to substantial. Erythromelalgia: In 1878, the Neurologist S. Weir Mitchell described a disease in which patients experience redness of their feet or hands with excruciating pain, which patients describe as being on fire. He strung
VIII. NEUROSCIENCE JARGON
18. “Neuro”-Dictionary
the Greek words for red (erythros), limb (melos), and pain (algos) together and termed the condition erythromelalgia. Inherited erythromelalgia remains a very rare pain syndrome, and many physicians will never encounter a patient with it. However, it is one of only a few pain disorders that has a clearly defined molecular and genetic basis, namely mutations in the gene encoding for the Nav1.7 sodium channel. Evidence-based medicine: Evidence-based medicine is the term used to describe the philosophical shift in medicine whereby treatments and interventions should be based on solid evidence provided by rigorous unbiased clinical studies rather than on the personal experience of any physician or anecdotal evidence. This philosophy implies that physicians have a responsibility to stay informed on the current state of drug discovery and clinical research in their respected fields, and that they will alter their practice and prescribing behavior as evidence in support of certain drug treatments or intervention changes. Excitation-inhibition balance (E-I balance): The E-I balance between glutamatergic and GABAergic synaptic activation determines the overall excitability of a neural network, and is presumably perturbed in epilepsy and other neurological conditions. Excitotoxicity: Excitotoxicity occurs when a cascade of processes that result in neuronal cell death is initiated through overactivation of NMDA-type glutamate receptors. These flux excessive amounts of Ca2 + into cells, which in turn activates destructive enzymes such as caspases, proteases, endonucleases, phospholipases, and others that destroy the cell and its DNA. Executive function: Executive functions are important mental processes that enable planning, strategizing, organizing, paying attention, remembering details, and managing time, resources, and space. Experimental autoimmune encephalomyelitis (EAE): The inoculation of experimental animals, mainly mice, rats, or rabbits, with components of myelin can elicit an autoimmune response called EAE that mimics many aspects of multiple sclerosis (MS). Therefore, EAE has become a popular model to study MS in laboratory animals. Extracellular matrix: The space surrounding brain cells is called the extracellular space and is filled with a collection of molecules called the extracellular matrix. These molecules include laminin, vitronectin, collagen, and hyaluronic acid. Extracellular matrix provides structural and biochemical support for the surrounding cells. Its gel-like consistency serves as a mechanical cushion but is also the substance through which cells signal to each other, for example, through growth factors. Following injury or in disease, extracellular matrix is degraded by proteases to allow for wound healing. Afterward, new extracellular matrix is deposited by adjacent cells.
443
Extracellular signal-regulated kinase (ERK)/mitogenactivated kinase (MAPK): Mitogen-activated kinase (MAPK) and extracellular signal-regulated kinase (ERK) are signaling molecules that add phosphate groups to proteins, thereby turning them on or off. These signals are engaged following the binding of an extracellular mitogen to a membrane receptor. An example of this is the epidermal growth factor binding to the epidermal growth factor receptor. Extra-pyramidal side effect: Extra-pyramidal side effects are attributable to indirect drug effects on nerves that do not travel through the “pyramids” of the brain stem. The pyramids harbor the corticobulbar and corticospinal tract, and are responsible for voluntary movements of extremities and cranial nerve innervation of the face, head, and neck. Improper use of antipsychotic drugs or narcoleptics can affect dopaminergic neurons in the basal ganglia, causing parkinsonian-like motor symptoms, such as bradykinesia or tremor. They can also affect dopaminergic neurons involved in emotions, leading to anxiety, distress, or paranoia. F-DOPA (fluorodopa): F-DOPA is an l-DOPA precursor radiolabeled with a radioactive 18-Fluor isotope. This allows for imaging of l-DOPA in the brain by positron emission tomography (PET). Fasciculations: Spontaneous involuntary muscle twitches are also called fasciculations. They can occur in people without disease, but may indicate a denervation of a small muscle group by nerve damage. Febrile seizures: Convulsive seizures that are evoked by elevations in body and brain temperature, typically through fever, are referred to as febrile seizures. Febrile seizures are common in infants and young children, who typically outgrow them. They are often associated with a loss of consciousness, and typically stop once the fever is controlled. Floppy baby syndrome: A loss of muscle tone caused by intoxication with botulinum toxin, floppy baby syndrome suggests that the baby has Clostridium botulinum bacteria resident in the intestines. Fluid percussion model: The fluid percussion model induces a reproducible form of brain injury that mimics closed head injury. The brain is pressurized through a cannula that is sealed to the skull and contains fluid that can be pressurized. The pressure of this fluid upon the brain causes damage. Fluor-deoxy-glucose (FDG): FDG is an analog of glucose that bears a radioactive fluor isotope, 18-fluor. It can be used as an imaging agent to measure glucose consumption in humans using positron emission tomography (PET). Glucose consumption is a surrogate measure for brain activity. Focal ischemic strokes: Strokes in which only a discrete brain region is affected by a blocked artery, without bleeding, are called focal ischemic strokes.
VIII. NEUROSCIENCE JARGON
444
18. “Neuro”-Dictionary
Focal seizures: A focal seizure originates in one brain hemisphere and typically affects only a small part of the brain. These were previously called partial seizures. They can present with and without dyscognia (loss of cognition) and, depending on the affected brain region, may present with convulsive muscle contractions affecting different parts of the body. Forced swim test (FST): The FST is a commonly used test of antidepressant efficacy that evaluates behavioral abnormalities in rodents. A mouse or rat is placed in a water tank from which it cannot escape. The animal is then observed for swimming behaviors and/or immobility (floating). Animals with a depressive-like phenotype will give up more quickly than normal animals, and thus rapidly adopt an immobile floating response. Commonly utilized antidepressants increase swimming behavior or decrease floating behavior, and thus this test is frequently used to screen new compounds for antidepressant efficacy. Fornix: The fornix is a fiber bundle that connects the hippocampus, thalamus, and nucleus accumbens. Fragile X: Fragile X is the most common genetic form of intellectual disability. It is caused by an abnormal trinucleotide expansion on the X chromosome adjacent to the FMRP protein. The silencing of the FMRP gene, which encodes an RNA-binding protein that regulates protein biosynthesis, causes an aberrant production of proteins that lead to the symptoms of fragile X. In addition to the neurocognitive impairments, fragile X patients show an unusually elongated face and ears and large testes. They are hypersensitive, are impulsive, and suffer attention deficits. Autistic features are also common. Fragile X mental retardation protein (FMRP): Fragile X mental retardation protein (FRMP) is an RNA-binding protein that regulates the translation of mRNA into protein. It is under the control of metabotropic glutamate receptor 5 (mGluR5). Normally, FMRP acts as a repressor of protein synthesis. Its absence causes protein biosynthesis to run amok. This is seen in fragile X, where the synthesis of the FMRP protein is reduced due to an instable region on the X chromosome adjacent to the FMRP gene, resulting in altered protein synthesis. Freud’s adjuvant: Freud’s adjuvant is a mixture of mineral oil and bacterial exotoxin containing membranes that stimulate a cell-mediated immune response. When used in combination with antigens such as myelin proteins to inoculate laboratory animals, it enhances the immune response. Frontotemporal dementia (FTD): FTD is a progressive neurodegenerative disorder that primarily affects the frontal lobe. It is the leading cause of dementia in patients under 65. FTD causes profound personality changes, including a loss of executive function and often socially unacceptable behavior. Language articulation is often impaired, while language comprehension remains intact.
Fused in sarcoma (FUS): Fused in sarcoma is a multifunctional DNA/RNA regulator that can bind and modify both DNA and RNA. Loss of function of FUS causes abnormal synaptic spine morphology, presumably by affecting specific synaptic proteins. Mutations in FUS account for 6% of familial cases of ALS. G protein-coupled receptors (GPCRs): GPCR is an umbrella term for membrane-associated receptors that respond to extracellular signals. GPCRs translate this extracellular signal to changes in signaling events inside the cells via intermediate G proteins or guanosine nucleotide-binding proteins. The resulting action can be inhibitory or excitatory. GABA hypothesis: The GABA hypothesis tries to explain the symptoms of schizophrenia through an impairment of GABAergic signaling, particularly a reduction in GABA content and density of GABAergic neurons. This in turn upsets the excitation-inhibition balance in the dorsolateral prefrontal cortex circuitry presumed to be critical in controlling proper association of different memory traces that form a narrative. Gamma oscillations: Gamma oscillations are high- frequency, 30–200 Hz synchronized voltage changes that can be recorded by EEG or extracellular electrodes in neuronal networks. Gamma oscillations are due to many neurons firing in synchrony, which is thought to bind the information processed in different parts of the brain together into a cognitive narrative or memory trace. For example, remembering a friend telling you a story may entail the recall of a memory with visual, auditory, and speech language information. This memory would therefore require synchrony between neurons in the visual cortex, hippocampus, auditory cortex, and speech language areas. In schizophrenia, incorrect elements of such narrative may be bound together due to alterations in gamma oscillations, resulting in hearing voices that are attributed to the wrong source. Ganglioside(s): These membrane-associated molecules have a long glycosphingolipid chain conjugated with sialic acid. These molecules were first isolated from the membranes of ganglion cells in the brain, hence their name. Gap junction: A gap junction is a channel-like connective pathway between adjacent cells. This pore allows for the exchange of small molecules